Pesticides in tap water: regional variations do not alter the findings of the national risk assessments

The term "pesticide" refers to substances or preparations for the prevention, control or elimination of organisms regarded as undesirable (plants, insects, fungi, rodents, bacteria, etc.).This term therefore covers not only plant protection products used in agriculture, but also biocides and certainhuman or veterinary antiparasitic drugs. Depending on the conditions of use and the characteristics of the environment, these pesticides can potentially be found in the environment(air, water, soil) and in food, including water. However, tap water is differentiated from other foods due to the dependence of individuals on a single source of supply and the geographical and temporal variability in contamination of this source.

What contribution does tap water make to total dietary exposure to pesticide residues? Has any regional variation been observed? What are the risks associated with drinking tap water?

These are the questions to which ANSES wished to respond. For this purpose, a study was conducted by the Agency, in the context of the missions of the French Observatory for Pesticide Residues (ORP), with funding from the French National Agency for Water and Aquatic Environments (ONEMA) as part of the Ecophyto plan, coordinated by the Ministry of Agriculture.

A study based on a very large number of tests

The aim of this study was primarily to assess the population’s dietaryexposure to pesticides from tap water and, on this basis, to characterise the risk associated with exposure via ingested water.

This work was based on all the data from the health inspections conducted by the Ministry of Health for the period 2007-2009, i.e. more than 5.7 million tests extracted from the SISE-Eaux health and environment information system on water. These tests cover abou t80,000 samples and concern 501 pesticide residues[1].

This study therefore supplements the results of the Second Total Diet Study (TDS2) and annual ANSES/ORP opinions on the monitoring of dietary exposure and definition of national food monitoring programmes. It is complementary because it includes geographic variability in water supply contamination.Contamination is thus considered at each sampling station and for each of the samples taken.

Among the 501 residues screened for in the health inspections:

• 210 were not quantified in the water supply (absent or present at levels too low to be quantified;).
• For 33 substances: it was not possible to reach any conclusion due to data deemed insufficient or missing.

The remaining 258 residues represent 224 substancesor groups of substances.

​​​​​​​For these 224 pesticides:

• the analysis focused as a priority on 106 pesticides for which data on contamination in water and solid foods are available ;
• for the remaining 118 substances, given their use, the previous risk assessments and their physico-chemical characteristics, they are rarely or never screened for in solid foods, and could not therefore be included in the estimation of water’s contribution to total dietary exposure. This is the case, for example, with chlortoluron, bentazone and glyphosate.

Based on these data, the share of exposure associated with water was calculated and compared with the exposure from solid foods to calculate the total dietary exposure. For each substance, this value was then compared with the toxicity reference values (ADI[2] for the long-term risk, ARfD[3] for the short-term risk) in order to characterise the risk.

Findings and outlook from this study

For the 106 substances for which it was possible to make an overall assessment at national level (in water and solid foods), the average contribution of water to total dietary exposure is less than 5%, except for eight pesticides and their metabolites: atrazine, simazine, oxadixyl, propoxur, benalaxyl, metolachlor, diuron and hexaflumuron.

The contribution of water is closely linked to the uses of the pesticides. In a few cases, it is substantial or even accounts for the total volume, for some herbicides that are found predominantly in water.

For some substances, variability in the exposure via water was observed within the same region, which could be explained by the diversity of cropping practices.

Regarding the long-term (chronic) risk, the study shows that water’s contribution to the acceptable daily intake (ADI) is low; it is less than 1%, except for two substances and theirmetabolites: atrazine and carbofuran which are now prohibited. For these substances, the contributionto the ADI is less than 5%.

The results of this study, taking into account the geographical variability in water supply contamination,do not therefore substantially alter the findings of the previous risk assessments (TDS2). The ADIs were exceeded for only two substances: dimethoate, which had already been identified in the TDS2 study, and propargite, which is now prohibited.

Regarding the short-term (acute) risk, there were no observed cases of toxicity reference values (ARfDs) being exceeded.

The results of this study

• the result of this study require no changes to the conditions for establishing the maximum health values proposed by ANSES at the request of the Ministry of Health in the event of non-compliance of tap water;
• the result of this study will provide input for the work carried out annually by ANSES/ORP for defining monitoring programmes for pesticides in food, particularly for substances that are rarely or never screened for in these matrices.

This work enables ANSES to formulate several recommendations, in particular:

• work to monitor the water supply should be maintained or even enhanced in small distribution units and for substances for which no conclusions could be reached in this study;
• in addition, further research is needed. This would help improve understanding of the health effects of certain pesticides and estimate the cumulative effects and effects at low doses of endocrine-disrupting pesticides.