Health Canada 2021 — Guidelines for Canadian Drinking Water Quality: Aluminum

Health Canada’s 2021 Guideline Technical Document for Aluminum in Drinking Water, prepared with the Federal-Provincial-Territorial Committee on Drinking Water, establishes a maximum acceptable concentration (MAC) for total aluminum of 2.9 mg/L (2 900 µg/L) and an operational guidance (OG) value of 0.100 mg/L (100 µg/L). The MAC is a health-based value derived from a NOAEL of 30 mg Al/kg bw per day for neuromuscular effects in the Poirier et al. (2011) developmental neurotoxicity study in rats exposed to aluminum citrate; the OG value is set to limit accumulation and release of aluminum and co-occurring contaminants (As, Cr, Mn, Ni, Pb, Cu) in the distribution system and to protect treatment and infrastructure performance. The document also compiles 2012–2017 occurrence data for total aluminum across Canadian municipal and non-municipal supplies in 10 provinces/territories, summarizes Canadian dietary exposure across seven age categories, and reviews the analytical, treatment and distribution-system controls used to manage aluminum in drinking water.

Key numbers

Guideline values:

• Maximum acceptable concentration (MAC) for total aluminum: 2.9 mg/L (2 900 µg/L), based on a locational running annual average of at minimum quarterly samples in the distribution system.
• Operational guidance (OG) for total aluminum: 0.100 mg/L (100 µg/L), based on a locational running annual average; intended to minimize aluminum accumulation/release, co-contaminant release, watermain deposition, and distribution-system disturbances.

Derivation of the health-based value (Section 3):

• Critical study: Poirier et al. (2011), Sprague-Dawley rats, aluminum citrate in drinking water at 0, 30, 100 and 300 mg Al/kg bw per day, in utero through lactation and post-weaning to 1 year of age, GLP/OECD TG 426 design (80/sex/group; 160 per dose group).
• Critical endpoint: developmental neuromuscular deficit (fore- and hind-limb grip strength, foot splay) plus clinical biochemical changes and renal pathology in male pups.
• NOAEL = 30 mg Al/kg bw per day (target dose); LOAEL = 100 mg Al/kg bw per day.
• Uncertainty factor = 100 (×10 interspecies, ×10 intraspecies).
• TDI = 30 ÷ 100 = 0.30 mg/kg bw per day.
• HBV = (0.30 mg/kg bw per day × 74 kg × 0.20) ÷ 1.53 L/day = 2.9 mg/L, where 74 kg = adult body weight, 0.20 = floor allocation factor for drinking water (food is dominant route at 95–99% of intake), 1.53 L/day = Canadian adult drinking water intake.

International comparison (Section 6):

• US EPA, EU, and Australia (NHMRC) have not set health-based limits; they have set aesthetic/operational guidance in the range 0.05–0.2 mg/L.
• WHO (2010) health-based value: 0.9 mg/L (non-regulatory), with practicable level emphasis at 0.1–0.2 mg/L; built on JECFA 2007 PTWI of 1 mg Al/kg bw per week, since revised to 2 mg/kg bw per week (JECFA 2012) using the same Poirier et al. (2011) key study Canada adopted.

Canadian dietary exposure (Table 1, µg total Al/kg bw per day):

Drinking water contributes ≤5% of total aluminum intake (range 0.8–4.4% across age groups; both extremes are in the 0–6 month group). The infant 0–6 mo breastfed drinking-water value of 0 represents direct ingestion only and excludes aluminum transferred via breast milk. The drinking-water column uses 111 µg/L as a conservative concentration estimate drawn from the surface-water distribution data in Table 2.

Canadian occurrence (Table 2; total aluminum, mg/L, 2012–2017 monitoring submissions):

National weighted means (Canada total, mean concentration, n/c indicates other statistics not calculated):

• Municipal ground-treated: 0.008 mg/L mean
• Municipal ground-distribution: 0.015 mg/L mean
• Municipal surface-treated: 0.120 mg/L mean
• Municipal surface-distribution: 0.111 mg/L mean

Jurisdictional highlights (median / mean / 90th percentile / max, mg/L):

• Newfoundland & Labrador, surface-distribution: 0.070 / 0.129 / 0.280 / 6.660 (n = 2 820/3 178 detections); non-municipal ground: 0.005 / 0.039 / 0.057 / 3.400.
• Nova Scotia, surface-distribution: 0.025 / 0.078 / 0.110 / 5.700 (n = 197/204); surface-treated: 0.048 / 0.086 / 0.177 / 0.724.
• New Brunswick, surface-distribution: 0.018 / 0.027 / 0.052 / 0.300 (n = 338/391).
• Quebec, ground-treated: 0.020 / 0.020 / 0.034 / 0.037 (n = 1/2); surface-treated: 0.029 / 0.084 / 0.200 / 0.360 (n = 6/6).
• Ontario, ground-and-surface-distributed: 0.024 / 0.042 / 0.109 / 1.340 (n = 1 212/1 387).
• Manitoba, surface-raw: 0.133 / 0.407 / 0.664 / 32.400 (n = 392/413); non-municipal ground: 0.002 / 0.010 / 0.014 / 0.266 (n = 51/144).
• Saskatchewan, non-municipal ground: 0.003 / 0.023 / 0.011 / 14.000 (n = 1 983/4 128); ground-and-surface-treated: 0.011 / 0.106 / 0.272 / 2.030 (n = 288/293).
• Alberta, surface-raw: 0.180 / 0.626 / 1.732 / 6.200 (n = 147/148).
• British Columbia, non-municipal not-specified: 0.020 / 0.059 / 0.050 / 3.000 (n = 313/352).
• Yukon, ground-treated: 0.005 / 0.005 / 0.005 / 0.025 (n = 11/68).

Method detection limits across the provinces ranged from 0.0002 to 0.409 mg/L. Mean and 90th-percentile concentrations in municipal surface water (treated and/or distributed) can exceed the 0.100 mg/L OG value; maximum concentrations in non-municipal supplies and municipal surface water can exceed the 2.9 mg/L MAC.

Canadian food data (Total Diet Study 2008–2012, Health Canada 2016): highest aluminum (>10 µg/g) in herbs and spices, baking powder, baked goods, processed chicken products, and chewing gum. Infant formula: milk-based 0.040–0.171 µg/g; soy-based 0.258–0.476 µg/g.

Bioavailability: oral aluminum absorption is approximately 0.3% (Stauber 1999; Krewski 2007), with chemical-form-dependent range 0.01–0.65% in humans and 0.01–5.1% in animals. Aluminum citrate (the form used in the key study) was chosen because it is the most soluble and most bioavailable form, making the derived HBV conservative across speciation. ~95% of absorbed aluminum is eliminated renally; brain elimination half-life in rats spans 13–1 635 days.

Acute exposure benchmark: the 1988 Camelford (UK) accidental contamination event exposed up to 20 000 people to aluminum concentrations 500–3 000× the WHO aesthetic value of 0.2 mg/L, producing nausea, vomiting, mouth and skin ulcers, and arthritic pain; follow-up studies did not demonstrate conclusive long-term effects (Lowermoor Incident Health Advisory Group 1989; McMillan 1993; Altmann 1999; UK Committee on Toxicology 2013).

Other animal points of departure considered: LOAEL of 1.5 mg Al/kg bw per day for neurobehaviour and for reproductive effects in Martinez et al. (2017a, 2017b); LD50 range 222–980 mg Al/kg in rats and mice.

Methods (brief)

This is a federal-provincial-territorial guideline technical document, not a primary research study. The document synthesizes (i) Canadian aluminum exposure data from the 2010 Priority Substances List Assessment Report (Environment Canada and Health Canada 2010) updated with 2012–2017 provincial/territorial drinking-water monitoring submissions, the National Drinking Water Survey (Health Canada 2017), and Environment and Climate Change Canada (2017); (ii) human and animal toxicology published through 2017, focused on neurotoxicity, neurobehaviour, and reproductive/developmental endpoints; and (iii) treatment, distribution-system and analytical literature relevant to managing aluminum in finished water.

Analytical methods reviewed for total aluminum in drinking water (Table 4): US EPA 200.5 Rev. 4.2 (axial ICP-AES, MDL 2.2 µg/L); US EPA 200.7 Rev. 4.4 (ICP-AES, MDL 20 µg/L); US EPA 200.8 Rev. 5.4 (ICP-MS, MDL 1.0–1.7 µg/L); US EPA 200.9 Rev. 2.2 (graphite furnace AAS, MDL 7.8 µg/L); APHA SM 3111D/3111E (flame AAS, MDL 100 µg/L); SM 3113B (electrothermal AAS, MDL 3 µg/L); SM 3120B (ICP-AES, MDL 40 µg/L); SM 3125 (ICP-MS, MDL 0.03 µg/L); SM 3500-Al B portable colorimetric (eriochrome cyanine R, 535 nm, MDL 6 µg/L); SM 3500-Al E online colorimetric (pyrocatechol violet, 580 nm, MDL 7–10 µg/L). Total aluminum requires acidification of unfiltered samples; dissolved aluminum requires filtration through 0.4–0.45 µm and acidification to pH <2 at the time of collection per APHA SM 3030B.

Risk-assessment derivation: Health Canada selected Poirier et al. (2011) over Martinez et al. (2017b) because Poirier was a large GLP/OECD TG 426 blinded study with multiple doses and multiple endpoints; the lower point of departure in Martinez (1.5 mg Al/kg bw per day) would have required additional uncertainty factors (LOAEL-to-NOAEL, subchronic-to-chronic) yielding excessive total uncertainty. Benchmark-dose modelling of Poirier was considered inappropriate because, for males, only two doses plus the control survived the high-dose mortality, and dose received varied with water consumption across study stages; the NOAEL of 30 mg/kg bw per day was therefore retained as the point of departure. A multi-route exposure assessment (skin, inhalation) was not performed because aluminum has low volatility and low dermal absorption (Flarend 2001; Pineau 2012).

Implications

Certification: Provides a regulator-derived, peer-reviewed Canadian point of departure for chronic oral aluminum exposure (NOAEL 30 mg Al/kg bw per day; TDI 0.30 mg/kg bw per day) that aligns with the JECFA 2012 PTWI revision and is anchored in a GLP developmental neurotoxicity study. The document also supplies Canadian occurrence data for drinking water as a context source (≤5% of total aluminum intake) against which food-side aluminum contributions can be benchmarked.

Courses: A useful worked example of how a national agency derives a drinking-water guideline from an animal NOAEL — including selection between competing key studies, choice of uncertainty factors, allocation factor floor logic, and the distinction between health-based MAC and operational/aesthetic OG values. Also documents the trade-off between aluminum-based coagulation (used in 69.2% of Canadian surface-water treatment per Statistics Canada 2013) and residual aluminum management.

App: Drinking water can be added to the per-age aluminum exposure model using the 0–4.4% allocation range from Table 1; the document supplies median, mean, 90th-percentile and maximum total-Al concentrations by water type and jurisdiction (Table 2) that can be used for region-specific exposure estimates.

Wiki pages this source may touch

• aluminum
• efsa-aluminium-twi — international comparator (EFSA TWI), referenced for comparison
• atsdr-aluminum-mrls — international comparator (ATSDR MRLs), referenced as a supporting risk-assessment source

Verification notes

• Source type is regulatory-guideline: this is a Health Canada Federal-Provincial-Territorial guideline technical document (March 2021), not a peer-reviewed primary research study. ISBN 978-0-660-37115-3, Cat H144-13/18-2021E-PDF, Pub 200347. No DOI. License: Government of Canada reproducible for personal/internal use with attribution.
• matrices: [drinking-water]. The document covers Canadian municipal and non-municipal drinking water, including treated, distributed, surface and groundwater types; the drinking-water umbrella slug is used per Part 5b routing-layer fan-out guidance.
• products: [] is correct: the document describes drinking water as a matrix and includes regulatory occurrence data; it is not a product-category study, and there is no provisional or scaffolded product slug needed.
• ingredients: [] is correct: the document describes aluminum exposure in food categories (Total Diet Study) only at a synthesis level (highest concentrations in herbs/spices, baking powder, baked goods, processed chicken, chewing gum, infant formula by category) and does not perform primary ingredient-level analysis.
• jurisdictions: [CA] covers all 10 sampling provinces/territories listed in sampling_locations. Canadian national-level guideline.
• metals: [Al] follows Part 14 abbreviation discipline. Aluminum is reported as total aluminum (dissolved plus particulate); no speciation distinction is drawn at the guideline level.
• Regulatory-slug proposal (for Karen): a new regulations/health-canada-aluminum-mac.md (or similar) slug appears justified for this document because it sets a binding Canadian MAC of 2.9 mg/L for total Al in drinking water. Per skill policy this skill does not create new regulation pages; surfaced here for the next regulation-slug review. Existing aluminum-regulation slugs are efsa-aluminium-twi and atsdr-aluminum-mrls only.
• Part 12 brand firewall: source contains no brand-attributed contamination data; method/vendor exception N/A (no analytical instrument vendors named in the source-level method summaries; APHA/EPA standard method numbers only).
• Part 2 wiki/HMTc firewall: this page reports the Health Canada guideline values, derivation steps, occurrence data, and international comparators as the document presents them. No comparison to HMT&C thresholds, no consumer risk advisories, no proposals for HMT&C certification levels.
• Brand-firewall exception 1 (regulatory event subject) does not apply — the document does not name any branded products as the subject of a regulatory action.
• Audit subagent (2026-06-04) flagged a Manitoba/Saskatchewan jurisdictional mis-attribution in the Canadian occurrence block: the non-municipal-ground row 0.003 / 0.023 / 0.011 / 14.000 (n = 1 983/4 128) was attributed to Manitoba but belongs to Saskatchewan per source Table 2 (page 5). Manitoba’s actual non-municipal-ground row is 0.002 / 0.010 / 0.014 / 0.266 (n = 51/144). Verified against source page 5 — finding correct; corrected.
• Audit subagent (2026-06-04) flagged the Poirier 2011 study size as (80 sex/group) being ambiguous (could be misread as 80 per group total). Source page 15 Table 3 explicitly states “80/sex/group” with the slash. Verified — finding correct; disambiguated to (80/sex/group; 160 per dose group).

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.