{"id":"france-drinking-water-quality","slug":"france-drinking-water-quality","name":"France Drinking Water Quality Dataset","description":"Annual drinking-water statistics by French department for five determinands — nitrate, lead, total pesticides, PFOS and arsenic — computed from the statutory sanitary control programme. Reports the exceedance rate, the mean quantified concentration and, unusually, the share of samples that were measurable at all.","methodology":"Computed by Hydrionis from per-sample results published by the regional health agencies through Hub’Eau. The governing decision is the treatment of censored results: French results below the laboratory detection limit are published with a textual result beginning \"<\" alongside a numeric field set to zero. Hydrionis treats the textual form as authoritative. Censored results are counted, published as a quantified-share series, excluded from every mean rather than substituted with zero or half the detection limit, and never counted as exceeding a limit. The resulting mean is therefore an upper bound and is labelled as one. Only complete calendar years are aggregated, and a department-year needs thirty results before a rate is published.","domains":["drinking-water","quality","contaminant"],"metricIds":["dw-nitrate-concentration","dw-nitrate-quantified","dw-nitrate-exceedance","dw-lead-concentration","dw-lead-quantified","dw-lead-exceedance","dw-pesticides-total-concentration","dw-pesticides-total-quantified","dw-pesticides-total-exceedance","dw-pfos-concentration","dw-pfos-quantified","dw-arsenic-concentration","dw-arsenic-quantified","dw-arsenic-exceedance","drinking-water-sample-count"],"subjectScope":"French departments, metropolitan and overseas","cadence":"annual","version":"1.0.0","released":"2026-08-25","changelog":[{"version":"1.0.0","date":"2026-08-25","summary":"First release. Hydrionis holds drinking-water observations for the first time.","additions":["Determinand statistics by department for five determinands."]}],"limitations":["A mean computed over quantified results only is biased upward. Where the quantified share is low — as it is for PFOS almost everywhere — the mean describes the detectable minority of samples rather than the water supplied, and must be read with that share.","Detection limits improve over time and differ between laboratories, so a rise in the quantified share can reflect better analysis rather than more of the substance.","Department-level statistics average over supplies that may differ greatly. A department with one contaminated small supply and a hundred clean ones will not look contaminated.","An exceedance in a department-year does not mean non-compliant water was supplied: regulators act on breaches, and the sample is what triggers that action.","Restricted to five determinands from a monitoring programme covering hundreds. Absence of a determinand here is not evidence about it."],"observationCount":4855,"seriesCount":1387,"subjectCount":101,"from":2022,"to":2025,"sourceIds":["fr-hubeau-eau-potable"],"units":["%","mg/l as NO₃","samples","µg/l"],"redistributable":true,"commerciallyReusable":true,"citation":"Hydrionis (2026), France Drinking Water Quality Dataset, version 1.0.0. Retrieved from https://hydrionis.com/data/datasets/france-drinking-water-quality/","downloads":[{"file":"/downloads/observations-drinking-water-fr-hubeau-eau-potable.csv","format":"CSV","label":"drinking-water observations from Hub'Eau / SISE-Eaux (5,171 rows)"}],"apiPath":"/api/v1/datasets/france-drinking-water-quality.json","lastFetched":"2026-10-01","fields":[{"id":"dw-nitrate-concentration","name":"Nitrate in drinking water","definition":"Mean concentration of nitrate, expressed as the nitrate ion in treated drinking water, computed over quantified samples only.","unit":"mg/l as NO₃","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A mean over quantified samples only. Where most samples are censored this describes the detectable minority, not the supply. Not comparable with a mean computed by substituting zero or half the detection limit for censored results, which is the more common convention elsewhere."},{"id":"dw-nitrate-quantified","name":"Nitrate quantified sample share","definition":"Percentage of drinking-water samples analysed for nitrate, expressed as the nitrate ion that returned a numeric result rather than a result below the detection limit.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Depends on the detection limit laboratories achieved, which improves over time. A rise can mean more of the substance is present or that it became visible at lower concentrations, and the two cannot be separated from this series alone."},{"id":"dw-nitrate-exceedance","name":"Nitrate limit exceedance rate","definition":"Percentage of drinking-water samples analysed for nitrate, expressed as the nitrate ion whose result exceeded the regulatory limit of 50 mg/l as NO₃.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Computed against the French regulatory limit of 50 mg/l as NO₃. Limits for the same substance differ between jurisdictions, so exceedance rates are not comparable internationally without checking both standards."},{"id":"dw-lead-concentration","name":"Lead in drinking water","definition":"Mean concentration of lead in treated drinking water, computed over quantified samples only.","unit":"µg/l","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A mean over quantified samples only. Where most samples are censored this describes the detectable minority, not the supply. Not comparable with a mean computed by substituting zero or half the detection limit for censored results, which is the more common convention elsewhere."},{"id":"dw-lead-quantified","name":"Lead quantified sample share","definition":"Percentage of drinking-water samples analysed for lead that returned a numeric result rather than a result below the detection limit.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Depends on the detection limit laboratories achieved, which improves over time. A rise can mean more of the substance is present or that it became visible at lower concentrations, and the two cannot be separated from this series alone."},{"id":"dw-lead-exceedance","name":"Lead limit exceedance rate","definition":"Percentage of drinking-water samples analysed for lead whose result exceeded the regulatory limit of 10 µg/l.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Computed against the French regulatory limit of 10 µg/l. Limits for the same substance differ between jurisdictions, so exceedance rates are not comparable internationally without checking both standards."},{"id":"dw-pesticides-total-concentration","name":"Total pesticides in drinking water","definition":"Mean concentration of the sum of the pesticides analysed in treated drinking water, computed over quantified samples only.","unit":"µg/l","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A mean over quantified samples only. Where most samples are censored this describes the detectable minority, not the supply. Not comparable with a mean computed by substituting zero or half the detection limit for censored results, which is the more common convention elsewhere."},{"id":"dw-pesticides-total-quantified","name":"Total pesticides quantified sample share","definition":"Percentage of drinking-water samples analysed for the sum of the pesticides analysed that returned a numeric result rather than a result below the detection limit.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Depends on the detection limit laboratories achieved, which improves over time. A rise can mean more of the substance is present or that it became visible at lower concentrations, and the two cannot be separated from this series alone."},{"id":"dw-pesticides-total-exceedance","name":"Total pesticides limit exceedance rate","definition":"Percentage of drinking-water samples analysed for the sum of the pesticides analysed whose result exceeded the regulatory limit of 0.5 µg/l.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Computed against the French regulatory limit of 0.5 µg/l. Limits for the same substance differ between jurisdictions, so exceedance rates are not comparable internationally without checking both standards."},{"id":"dw-pfos-concentration","name":"PFOS in drinking water","definition":"Mean concentration of perfluorooctane sulfonic acid in treated drinking water, computed over quantified samples only.","unit":"µg/l","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A mean over quantified samples only. Where most samples are censored this describes the detectable minority, not the supply. Not comparable with a mean computed by substituting zero or half the detection limit for censored results, which is the more common convention elsewhere."},{"id":"dw-pfos-quantified","name":"PFOS quantified sample share","definition":"Percentage of drinking-water samples analysed for perfluorooctane sulfonic acid that returned a numeric result rather than a result below the detection limit.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Depends on the detection limit laboratories achieved, which improves over time. A rise can mean more of the substance is present or that it became visible at lower concentrations, and the two cannot be separated from this series alone."},{"id":"dw-arsenic-concentration","name":"Arsenic in drinking water","definition":"Mean concentration of arsenic in treated drinking water, computed over quantified samples only.","unit":"µg/l","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A mean over quantified samples only. Where most samples are censored this describes the detectable minority, not the supply. Not comparable with a mean computed by substituting zero or half the detection limit for censored results, which is the more common convention elsewhere."},{"id":"dw-arsenic-quantified","name":"Arsenic quantified sample share","definition":"Percentage of drinking-water samples analysed for arsenic that returned a numeric result rather than a result below the detection limit.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Depends on the detection limit laboratories achieved, which improves over time. A rise can mean more of the substance is present or that it became visible at lower concentrations, and the two cannot be separated from this series alone."},{"id":"dw-arsenic-exceedance","name":"Arsenic limit exceedance rate","definition":"Percentage of drinking-water samples analysed for arsenic whose result exceeded the regulatory limit of 10 µg/l.","unit":"%","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"Computed against the French regulatory limit of 10 µg/l. Limits for the same substance differ between jurisdictions, so exceedance rates are not comparable internationally without checking both standards."},{"id":"drinking-water-sample-count","name":"Drinking water samples analysed","definition":"Number of regulatory drinking-water sample results held for a supply area over a period.","unit":"samples","nativeGeography":["supply-zone","local-authority","county"],"comparabilityNotes":"A count of results Hydrionis holds, which is the count the publisher exposed through its API. It is not a statement about how many samples a supplier was legally required to take."}]}