Pesticide authorisations put our ground-& drinkingwater at risk

A hand holding a water glass can be seen. Click to enlarge
Degradation products of pesticides can diminish groundwater and drinking water quality
Source: bigfoot /

The quality of the drinking water in Germany is excellent. Its main source, groundwater, enjoys a high level of protection and should be as free as possible from chemical residues.

Most plant protection products are applied directly into the environment. The active substances move into the soil matrix and degrade there. During this process, new substances are formed. Some of these are biologically less active and toxic than the actual active ingredient, but are associated with considerable problems for the environment and drinking water production: Many of these degradation products are very mobile and easily leach into groundwater. Numerous of such substances can already be found in German groundwater bodies - significantly more and in higher concentrations than their parent substances. Because many of these substances are difficult to remove, they are found in our drinking water. This is because groundwater is the most important source of drinking water in Germany (UMID article).

Inputs of some degradation products into groundwater may no longer be regulated

The inputs of these degradation products are limited by the plant protection product authorisation: If inputs to groundwater are expected to exceed 10 micrograms per litre (µg/L) as a result of product use, the respective product could not be authorised in Germany so far. However, in two recent cases, the courts have decided that the authorisation must nevertheless be granted according to the current legal situation. However, we fear that this decision is going to establish a principle that lowers the level of protection for our groundwater and drinking water - if the legislator does not take countermeasures.

In soil, the maize herbicide S-metolachlor breaks down into various degradation products, some of which leach into the groundwater in concentrations far above 10 µg/L. The fact that the UBA opposed the authorisation of a product containing S-metolachlor in Germany was found to be unlawful by the court. This was justified by the worksharing in the authorisation procedure between the EU states: According to this, one state - chosen by the pesticide manufacturers themselves - is initially responsible for evaluating the product. If the risks associated with the use of the product are considered to be sufficiently low, it can then be authorised in other states without an individual assessment (see info box below). However, S-metolachlor applications have a critical impact in Germany in particular, because precipitation is relatively high and the soil is very permeable in some regions. Some degradation products are already frequently found in groundwater, sometimes above the Health Related Indication Value (HRIV) of 3 µg/L. The HIRV is a technical recommendation of the Federal Environment Agency (more about HRIV), which most health authorities set as a binding limit: If the HRIV is exceeded, the water supplier may be forced to carry out cost-intensive water treatment. Because of the contamination with degradation products of S-metolachlor, some Bavaria classified three groundwater bodies in "poor chemical status" - an EU criterion for assessing groundwater quality.

The weed killer flufenacet degrades in the soil to trifluoroacetate (TFA), which also leaches into the groundwater in large quantities and is not degraded further (TFA background paper). Binding requirements that restrict the amounts and frequency of flufenacet application can limit inputs to below 10 µg/L. However, according to the court, these conditions are not compatible with the EU authorisation procedure - because the EU state in charge of the assessment used outdated data and did not even take the degradation product TFA into account (see info box below). The UBA had taken into account the high expected groundwater inputs, proven water pollution and already existing conflicts with drinking water production in its decision on whether to grant authorisation. However, as in the case of S-metolachlor, the court focused on the division of labour among EU member states, which severely restricts the scope of action of the individual EU states - and judged the conditions to be unlawful (UBA article "Pesticide authorisations undermine environmental protection"). What the court had decided for a single product was subsequently applied to all other eleven plant protection products containing flufenacet. TFA inputs from flufenacet into groundwater - for most applications well above 10 µg/L - are thus no longer restricted. TFA is already detected at 81 % of all groundwater monitoring sites, with local levels above 10 µg/L. Concentrations are likely to rise in the future - not only because TFA is persistent in groundwater, but also because flufenacet application is on the rise (BVL sales figures).

Less toxic does not mean unproblematic

If the degradation products are significantly less biologically active and toxic - what is the problem though? Significantly less does not mean that risks can be completely ruled out, especially since some of the inputs into groundwater are incomparably higher than those of active substances. Since degradation products have always been treated rather neglectfully in the assessment, we know little about them. For them, far fewer studies need to be submitted than for active substances - on behaviour in the environment, effects on ecosystems and health effects, for example. It is not uncommon for them to be classified as harmless, only to discover a few years later that they have an effect on humans and animals. Only recently, for example, it was revealed that another degradation product of the aforementioned S-metolachlor still has a high residual biological activity. A stricter limit value now applies to it, which is already frequently exceeded in groundwater and drinking water.

Active substances are developed in a way that they quickly degrade in the soil to substances that are harmless to health - in principle a good idea. However, it often follows that such substances themselves decompose very poorly and are also very mobile. Some tend to accumulate in groundwater. What if health risks are identified for individual substances or for their mixture in groundwater? Most of these substances cannot be removed with the usual methods of drinking water treatment. The only option would be cost-intensive treatment by the water suppliers. In the case of TFA, a reverse osmosis system would have to be installed, which, however, also removes essential micronutrients. But even if the substances are not considered critical to health: Residues of chemicals should be kept within narrow limits in order to keep drinking water quality high in the long term. This is what the minimising principle demands as a principle of drinking water law (Drinking Water Ordinance, § 6, para. 3). Various water suppliers and water associations are already sounding the alarm and calling on politicians to restrict the use of pesticides in drinking water abstraction areas (article on Süddeutsche Zeitung).

The problem is likely to get worse

Water is becoming more and more precious in Germany as well: some communities already suffer from water shortages in summer, a situation that is likely to worsen with the expected effects of the climate crisis. Instead of tolerating ever higher concentrations and risks of chemicals in groundwater and drinking water or building expensive methods to remove them, inputs should be kept as low as possible. Here, plant protection product authorisation must also play its part. However, the cases of S-metolachlor and flufenacet have shown that, on the basis of current law, the level of protection of groundwater and drinking water in Germany is decreasing. Other substances may follow. Based on the pesticides authorised in Germany, there are about 300 degradation products that can enter groundwater in significant quantities. Such substances have been detected at 58 % of groundwater monitoring wells in Germany. This is only the tip of the iceberg, because most degradation products are not yet part of the groundwater monitoring programmes (factsheet on degradation products of pesticides).

The degradation products are neglected mainly because there are no parametric limit values that are legally binding. This inconsistency in the legal framework provides a target for lawsuits by companies. The Federal Environment Agency urgently advises modernising the affected and outdated regulatory frameworks at national and European level - with the aim of creating consistent and binding legal regulation. The most effective lever is regulation at the source of entry, in the authorisation of plant protection products. A binding limit value for all degradation products in the Plant Protection Products Regulation (Regulation (EC) 1107/2009) could manage inputs uniformly across the EU. This limit value should consider not only toxicological effects but also environmentally critical properties such as persistence, mobility and risks for drinking water abstraction. In order to realistically assess the extent of pollution, significantly more degradation products would have to be monitored in the aquifers. In the recommendation list, the Federal Environment Agency proposes a concrete priorised list of substances to be measured.

According to the EU Commission, the use of pesticides in Europe is to be significantly reduced. Recently, it published a proposal for a new EU regulation that stipulates the reduction of pesticide use by half by 2030. The current authorisation of plant protection products obviously works against these goals. Therefore, the worksharing among EU states in authorisation should be reorganised in order to strengthen environmental protection across Europe. More on the new EU regulation here.

Approval and authorisation of plant protection products

Before plant protection products may be sold and used, they must pass a two-stage assessment process. First, the active ingredient is thoroughly investigated - for efficacy, environmental performance, ecological and health risks - and can finally be approved at EU level for 7-15 years.

Approval is necessary in order to use the active ingredient in products that farmers then apply to the fields. The products themselves - usually consisting of several active substances and co-formulants - go through an authorisation process in which the efficacy and the effects on human health, and the environment are evaluated. Formally, each EU state grants authorisation on its own, but a comprehensive evaluation is carried out by only one state, which the other countries - with very few exceptions - must follow. The aims of this regulation are an efficient division of labour and harmonised product authorisation in the EU to ensure the free movement of goods. However, manufacturing companies can choose the evaluating state themselves and thus take advantage of the different framework conditions for authorisation in the individual countries as well as gaps in the harmonised assessment methodology. This lowers the standard of environmental protection throughout the EU - more in the UBA article "Pesticide approvals leverage environmental protection".

The product evaluation is in turn based to a large extent on the results of the EU active substance approval. Since this is theoretically renewed every 7-15 years, the data basis for the product evaluation is also correspondingly up-to-date. In practice, however, we notice a delay in the re-approval of many active substances. The active substance flufenacet, for example, was last approved in 2004. The review was repeatedly postponed and has not been completed to date; we expect an official conclusion in 2023 at the earliest. For one product, the court had prohibited the use of more recent data than those from 2004, although they had shown various risks for the environment. Product authorisations with flufenacet are thus far from the "current state of science and technology" that the EU Plant Protection Products Regulation actually requires (Regulation (EC) 1107/2009, Art. 29 (1)e).