A2 Drinking Water Directive

A2.1      Introduction

Legislation

The DWD aims to ensure that water intended for human consumption is safe. It must be free of any microorganisms, parasites or substances that could potentially endanger human health. The directive applies to all water intended for human consumption apart from natural mineral waters and waters that are medicinal products.

The directive came into force in 1998 and replaced Directive 80/778/EEC. 27 Member States of the EU have enacted it in their national legislation and have to comply with its requirements (Croatia do not yet have to comply with the DWD).

The directive:

• sets quality standards for drinking water at the tap (microbiological, chemical and organoleptic parameters) and the general obligation that drinking water must be wholesome and clean;

• obliges Member States to monitor drinking water quality regularly, to take remedial action if the monitoring reveals problems and to provide consumers with adequate and up-to-date information on the quality of their drinking water;

• allows Member States to exempt water supplies serving fewer than 50 persons or providing less than 10 m³ of drinking water per day on average, drinking water from tankers, drinking water in bottles or containers and water used in the food-processing industry, where the quality of water cannot affect the wholesomeness of the foodstuff in its finished form.

Drinking water quality parameters

The directive sets standards for the most common organisms and substances that can be found in drinking water. A total of 48 parameters must be monitored and tested regularly. In general, the basis for the standards is the WHO’s guidelines for drinking water and the opinion of the EC’s Scientific Advisory Committee.

Annex I of the directive divides the parameters into microbiological parameters, chemical parameters, indicator parameters and radioactivity.

The two microbiological parameters are E. coli and enterococci. Their parametric value is a substitute for zero. In other words, these organisms should be absent from drinking water to guarantee its quality.

The indicator parameters are not directly relevant to the quality of water. They indicate that something has changed in the source, the treatment or the distribution of the water. This needs to be investigated and may require urgent action. Most indicator parameters do not pose a direct threat to human health, but they might have an indirect impact through the appearance, taste or odour of the water, making it less acceptable to the consumer, or they might interfere with proper treatment. For example, organic matter may make disinfection inadequate.

 The chemical parameters were selected for their potential impact on human health, and they do not closely match the list of priority substances under the WFD. Apart from accidents, chemicals are almost never present in drinking water in concentrations that cause acute health effects. They include trace elements, such as arsenic, nickel or lead, and other substances, such as cyanide, polycyclic aromatic hydrocarbons or nitrogen compounds (nitrate and nitrite). In particular, these parameters cause ‘blue baby syndrome’ (see Chapter 5). Furthermore, the impact of these chemicals depends on how they affect the human body. Mostly, the parametric values are based on lifelong exposure and an average intake of 2 l of drinking water per person per day. There is a distinction between threshold and non-threshold substances:

• Threshold chemicals have no impact on human health when concentrations are below the threshold. In cases of non-compliance, the impact depends on the amount of exceedance, the duration of exposure and the safety factor that has been used in setting the parametric value. This differs for each parameter, based not only on health impacts but also on technological capability and ability to analyse substances in the water.
• Non-threshold chemicals, such as pesticides, have no threshold below which there is no potential effect on human health. Dealing with them uses a risk approach that mostly accepts one additional death through drinking water among 1 million people; this is stricter than the value that the WHO currently uses (1 in 100 000 people). If we know the level of non-compliance and the duration, we can then try to estimate the potential impact on human health in a particular Member State or water supply zone (Hulsmann et al., 2015).

Member States may, for a limited time, deviate from the chemical quality standards specified in Annex I of the directive. This process is called ‘derogation’. A derogation can be granted if it does not constitute a potential danger to human health and if there is no other reasonable means of maintaining the supply of water intended for human consumption in the area concerned.

Drinking water quality must be checked at typical locations. They must represent the water source, treatment plant, storage facilities, distribution network, points at which water is delivered to the consumer and points of use. Therefore, where to sample the water depends on the parameter and the potential risk. For instance, excess lead mostly comes from domestic pipe systems and is not a problem in waterworks, so it is all the more important to sample water for lead at the tap, after it has gone through the distribution network.

Water supply zones and dependency on environmental pressures

Assessing drinking water quality is based on the spatial scale of a water supply zone. A water supply zone is ‘a geographically defined area within which water intended for human consumption comes from one or more sources and within which water quality may be considered as being approximately uniform’ (Annex II, DWD). This means that a water supply zone could be a waterworks that collects and processes raw water from two drinking water reservoirs; or it could also be an elevated tank that supplies a district with drinking water.

 The directive makes a distinction between large and small water supply zones. Large water supply zones supply more than 1 000 m³ of drinking water per day, on average, or serve more than 5 000 persons. Small water supply zones are subdivided into three further categories: category 1 supplies 10–100 m³ per day; category 2 supplies 100–400 m³ per day; and category 3 supplies 400–1 000 m³ per day.

The level of treatment required in a particular water supply zone depends on the quality of water it receives from its sources. Water efficiency measures are worth exploring if there are shortages of raw water or if polluted sources require costly treatment. The option to transfer water from another basin with abundant resources would need to be assessed with the financial and environmental costs of transfer against the costs of treatment and possibility to reduce pollution at source (for costs of transport see also e.g. Kraemer et al., 2007; Holländer et al., 2008). Measures to reduce demand and clean up pollution should be exploited before considering the need to transfer water between basins. The DWD itself does not request any information about the quality of raw water, its source (in terms of transfers from other water bodies) or the intensity of treatment necessary (see also Chapter 3). Integrated approaches to evaluate the most appropriate solution taking source water quality into account  be implemented in an integrated way under the WFD.

Information to the public

Large and small supply zones have the same minimum water quality requirements. However, monitoring requirements differ. Reporting to the EC is mandatory for large water supply zones. Member States are also obliged to report the water quality of small water supply zones if data are available.

Every three years, Member States must digitally report drinking water quality to the EC. They are also obliged to publish a national report to the public. Table A2.1 lists links to the national drinking water quality reports or information about the reporting period 2011–2013. Furthermore, national country reports show drinking water quality in the reporting period 2011–2013 in a nutshell (map viewer on dwd.etcicm.cenia.cz). All reported data are available in the WISE databases.

Table A2.1 Links to national drinking water reports and information for 2011–2013

A2.2      European results

General information

According to the drinking water quality data of reporting period 2011 to 2013 in the EU, the volume of water supplied divided by number of resident population is about 220 l. This is much more than the mean water consumption per person and day, because drinking water supply for industry or other uses than human consumption is included. The water consumption varies between Member States; for example it is about 81 l in Slovakia, 150 l or less in Denmark and Hungary, and more than 200 l in Austria, Bulgaria, Cyprus, Finland, France, Greece, Ireland, Italy, Luxembourg, Portugal, Romania, Slovakia, Sweden and the United Kingdom. The high consumption level also includes water for agriculture and/or industry in many cases. This water comes from various sources, mainly groundwater or surface water (e.g. drinking water reservoir). Overall, the main source in EU’s countries is groundwater, which provides some 50 % of the total (Figure A2.1). Figure A2.2 shows the percentages of water sources in EU Member States (before Croatia’s accession).

Figure A2.1 Sources of drinking water in the EU, 2011–2013.

Figure A2.2 Sources of drinking water, 2011–2013.

*In the Czech Republic, inland water is synonymous with surface water.

Figure A2.3 shows the percentage of resident population in large water supply zones (> 1 000 m³ per day and/or supplying more than 5 000 people). Some countries also reported the resident population of small water supply zones for in 2011–2013. If we add those, the total proportions of residents supplied are 100 % in Bulgaria, France, Hungary, Malta, Portugal and Slovakia; 96 % in Belgium; 95 % in Cyprus; 93 % in Spain; 90 % in Slovenia; 85 % in Ireland; and 66 % in Romania.

Figure A2.3 Population with access to large water supply zones, 2011–2013.

Drinking water quality

To assess drinking water quality in water supply zones, Member States carried out a huge number of analyses in the reporting period 2011–2013: microbiological parameters (4.1 million analyses), chemical parameters (7.1 million analyses) and indicator parameters (17.5 million analyses).

For each parameter, information on compliance was available. Percentage of compliance reflects the ratio between the number of analyses and the number of exceedances. Compliance with the Directive means that more than 99 % of all analyses meet the given standard. Exceeding indicator parameters does not necessarily mean non-compliance with the directive, for the reasons mentioned in section A 3.1, but only if there is no direct threat to human health.

Figure A2.4 shows the percentage of compliance for the parameter groups: microbiological, chemical and indicator parameters. The results show high compliance rates for microbiological and chemical parameters. Indicator parameters reached almost 99 % compliance in the reporting years 2011–2013. The indicator parameters covered exclude colour, odour, taste and turbidity, which do not have numerical values.

Figure A2.4 Percentage of compliance for microbiology, chemicals and indicator parameters in the EU.

Figure A2.5 shows information on compliance for the chemical parameters in the EU.

Figure A2.5 Compliance rates for the chemical parameters in the EU, 2011–2013.

At 98.83 %, arsenic shows the lowest compliance rate. This is mainly caused by geological background concentrations in the catchment areas.

Figure A2.6 shows information on exceedances for indicator parameters. The figure just gives an overview of the exceedances. It does not reflect non-compliance with the directive, because a number of indicator parameters do not have a numerical value, such as colour, taste, odour or turbidity.

The most frequent exceedances in the indicator parameter group are for total organic carbon, iron, sulphate and manganese.

Figure A2.6 Compliance rates for the indicator parameters in the EU, 2011–2013, excluding colour, taste, odour and turbidity.

Box A2.1 reflects the importance of pesticides as a potential risk for drinking water quality.

Box A2.1 Pesticides in drinking water

Pesticides can be ‘contaminants of concern’ for aquifer recharge, and mainly reach aquifers from agricultural run-off.

Much effort has been put into standards for pesticide levels in drinking water. The standards use an indicator approach and do not really reflect acceptable concentrations for health. The DWD sets a concentration limit of 0.1 μg/l for individual pesticides, and the sum of the pesticides must not exceed 0.5 μg/l. Because pesticides are present on a regular basis and in low concentrations, exposure to these chemicals is generally chronic. The health risk is difficult to assess, because data on acceptable doses for chronic exposure are scarce and the low concentrations involved are difficult to monitor.

Member States monitor a huge number of national specific pesticides and metabolites in drinking water. However, the EC and Member States agreed a short list of 13 pesticides for which the Member States reported monitoring frequency and non-compliance in 2011–2013. The short list is a harmonised approach and makes reporting comparable, but does not show the full picture of all pesticides and all relevant metabolites in a country.

The following figure shows the percentage of large water supply zones in the EU, where monitoring of the listed pesticides have been carried out in the reporting period 2011–2013.

Admitting that monitoring is low, compliance rates are consistently high. The compliance rate is more than 99.9 % for pesticides in total, but 99.6 % for individual pesticides (see Figure A2.5). These include the region-specific substances and all relevant metabolites.

Protecting raw water is particularly important. Critical groundwater bodies need special attention from specific measures for drinking water. That cannot be the task of the water suppliers alone. Rather, various stakeholders need to cooperate closely to plan and implement measures in the catchment area. To protect drinking water against pollution from the catchment area, there must be a well-integrated link between the DWD and the other water-related directives, such as the WFD and the BWD.

Causes of non-compliance

The reporting obligations mean that, if a water supply zone does not comply, Member States need to report causes and remedial actions. The number of causes depends on the number of non-compliant analyses.

Failures suspected of being caused by contamination of the source water are defined as ‘catchment related’. These causes include discharges from wastewater treatment plants or stormwater overflow (see also Chapters 3 and 6), agricultural activities (use of fertiliser and pesticides) and industrial activities. Treatment-related causes are mainly associated with the treatment processes at the plant, such as chemical dosing regimens, coagulation and clarification procedures, filter operation or disinfection. Within the distribution network, causes of contamination could be flow reversals and pressure changes, changes in the flushing or scouring regime, or leakages. Failures associated with the domestic distribution network can be identified only at the consumer’s tap. Non-compliances with the standards for copper, lead and nickel at the consumer’s tap may be associated with the consumer’s pipes and fittings.

Figure A2.7 shows the main parameters that exceeded the parametric value and that had causes reported. During the reporting period 2011–2013, most of the reported causes of exceedances were for coliform bacteria. A number of causes were also reported for iron, microbiological parameters other than coliform bacteria, total organic carbon, ammonium and manganese. Most of these are indicator parameters that pose no direct threat to human health.

Figure A2.7 Number of analyses with reported causes of exceedances for the parameters of the DWD in the EU, 2011–2013.

Figure A2.8 shows the different causes for the most reported parameters. Causes of exceedances due to biological parameters (coliform bacteria, colony count, E. coli, enterococci, clostridium) and iron cannot be exactly specified, exceedances of ammonium, manganese, pH, chloride, sulphate, arsenic and nitrite are mainly related to the catchment. Total organic carbon and aluminium mainly come from treatment, whereas lead is clearly associated with problems in domestic distribution networks. (So is nickel, which is not shown in the figure.)

The problem of nitrite and the nitrate/nitrite formula is widely discussed. The figure shows a rather small number of reported causes. They are obviously related to the catchment and are less prone to sudden impacts within the source area. Pre-selection and combination of higher-quality raw waters usually mitigate them.

Figure A2.8 Causes of non-compliance for the most reported parameters of the DWD in the EU, 2011–2013.

A2.3      Country comparison

Table A2.2 presents compliance of parameter groups for each EU Member State (before Croatia’s accession). The percentages are the mean compliance rate for each parameter group in 2011–2013.

Table A2.2 Compliance rates in 2011–2013 (%)

*Except odour, taste, colour and turbidity.

For the microbiological parameters, all Member States reported between 99 % and 100 % compliance. For the chemical parameters, 26 Member States reported compliance of between 99 % and 100 %, and only Hungary reported between 98 % and 100 % compliance.

For the indicator parameters, three Member States had compliance rates between 98 % and 100 %, three had compliance rates of less than 98 % and 21 had compliance rates of 99 % to 100 %. Malta reported a mean compliance rate of less than 98 % because of very low compliance rates for chloride.

A2.4      Measures to improve drinking water quality

Where the quality of drinking water needs further improvement, the action to take depends on the parameter and the cause.

Some parameters, such as nitrate or pesticides, relate to human activity in the catchment area. In these cases, measures need to be taken to improve site-specific source protection over a long time. For example, authorities could improve wastewater treatment plants or restrict the use of fertiliser and pesticides within the zone of contribution. They also need to liaise with the teams implementing the river basin management plan under the WFD. Short-term remedial actions could be additional treatment or changing the source of raw water; however, that might require longer water transfers.

For treatment-related parameters, remedial actions include changes in chemical dosing regime, coagulation, clarification procedures, filter operation (backwashing arrangements) or disinfection. Remedial actions related to distribution networks are, inter alia, flushing/scouring the mains or replacing/refurbishing corroded/leaking pipes.

If loss of drinking water quality is linked to the use of materials such as lead, copper and nickel, problems are often related to the domestic distribution network and in-house installation. The only ways to solve them are conditioning the water and informing the public about the proper use of materials and actions they can take to avoid too high levels of those substances in their drinking water.

Figure A2.9 shows the percentages of different kinds of remedial actions for three parameters that cause non-compliance or had a higher number of exceedances in 2011–2013: coliform bacteria, lead and total organic carbon.

Figure A2.9 Remedial actions for selected water quality parameters in the EU, 2011–2013.

Remedial actions for coliform bacteria contamination mainly affect the public distribution network or treatment. Remedial actions to minimise high concentrations of arsenic in drinking water are related to treatment (46 %) or catchment (29 %). For concentrations of lead exceeding the standards, 67 % of all reported remedial actions were replacing or disconnecting lead pipes in the domestic distribution network.

To summarise, specific drinking water quality parameters or groups of parameters cause problems at different points of extraction: water source, treatment, distribution and end of pipe — the consumer. That makes it difficult to develop transparent and useful monitoring, to identify the causes of non-compliance and to implement measures to maintain a healthy supply of drinking water in the EU.

A2.5      Information dissemination

Whereas up-to-date information about bathing water is available to the public at the bathing sites, through the media, in reports and through data viewers, data on drinking water are frequently not up to date and can be difficult to understand. Most Member States do not use comprehensive maps or other published public reports, except a yearly report on national drinking water quality. This makes it difficult to provide the public with updated EU-wide information on drinking water policy and quality on a regular basis. In addition, countries monitor, process and report data in different ways across the EU. That makes it difficult to compare the performance of different Member States and their compliance with the directive. A revision of the directive is now in progress. It is updating the monitoring requirements and methods of analysing drinking water quality parameters (Annexes II and III; DWD). Furthermore, the digital reporting is planned to be further revised and harmonised, based on the lessons learned from the reporting exercise for 2011–2013.