Fluoride

Adding a little is policy. Finding a lot is geology. The EPA's enforceable limit hasn't been updated for new evidence in a generation

EPA MCL

Primary MCL: 4.0 mg/L. Secondary (cosmetic): 2.0 mg/L. Optimal added (community fluoridation): ~0.7 mg/L.

Health concern

Dental fluorosis at chronic 2-4 mg/L; skeletal fluorosis at 8+ mg/L sustained; possible reduced IQ in children at chronic exposure above 1.5 mg/L (NTP 2024)

Testing method

Ion-selective electrode or ion chromatography; $15-25; included in basic well panels

Fluoride is the contaminant where the public conversation is dominated by the policy debate over adding a small amount to municipal water for cavity prevention (~0.7 mg/L), and almost nobody talks about the much larger problem of the natural fluoride that shows up in many private wells across the western US at levels well above the EPA's enforceable limit. If your municipal water is fluoridated, you can opt out by drinking RO water; if your well water has natural fluoride, you can't opt out without treatment.

The EPA's primary MCL of 4 mg/L was set in 1986 and hasn't been updated since. The scientific evidence has moved substantially since then, particularly the 2024 National Toxicology Program review which concluded with "moderate confidence" that chronic exposure above 1.5 mg/L is associated with reduced IQ in children. The MCL hasn't moved. The gap between regulatory standard and current evidence is one of the biggest in US drinking-water regulation.

Where natural fluoride comes from

Fluoride dissolves into groundwater from fluoride-bearing minerals: fluorite (CaF₂), fluorapatite, and trace fluoride in weathered volcanic and granitic rocks. Solubility is highest in alkaline, low-calcium waters — so the same aquifers that produce naturally elevated arsenic and uranium often also produce naturally elevated fluoride.

US hot zones for natural fluoride:

• Western Texas and the Texas Panhandle — substantial fractions of Ogallala wells in the southern High Plains exceed the 4 mg/L MCL.
• Eastern New Mexico — Pecos River Basin, Curry and Roosevelt counties.
• Arizona — particularly the Basin and Range valleys; Maricopa, Pinal, Pima counties have documented exceedances.
• Nevada and Utah — across the Basin and Range province.
• Colorado Western Slope and parts of the San Luis Valley.
• Idaho Snake River Plain — particularly in the McCall area and parts of the Magic Valley.
• South Carolina Coastal Plain — pockets in the Black Creek aquifer.
• North Dakota and parts of South Dakota — the Dakota aquifer.

Many of these regions have community water systems that exceed the 4 mg/L MCL and are required to defluoridate; their private-well neighbors typically don't.

Health effects: the moving target

The clinical effects of fluoride exposure scale with both concentration and duration:

• Dental fluorosis — staining and pitting of permanent teeth from chronic exposure during tooth development (roughly age 0-8). Mild fluorosis appears at sustained exposures around 1.5-2 mg/L; moderate-to-severe at 2-4 mg/L. Cosmetic, not functional, but visible.
• Skeletal fluorosis — bone abnormalities and joint damage from long-term exposures above ~8 mg/L sustained over years. Rare in the US (most exceedances are in the 2-6 mg/L range, not the 10+ mg/L range), severe in some endemic regions of India, China, and Africa.
• Cognitive effects — the 2024 NTP review concluded that chronic exposure above ~1.5 mg/L is associated with measurable IQ decrement in children. The mechanism is not fully established. The dose-response below 1.5 mg/L is uncertain.
• Thyroid effects — emerging evidence of mild thyroid suppression at chronic exposures in the 1-4 mg/L range.

The 0.7 mg/L community-water-fluoridation level was set on the assumption that the cavity-prevention benefit at that concentration outweighs the (then poorly characterized) low-level risks. The benefit is well-established. The risk side has gotten clearer in recent years.

Testing

• Method: ion-selective electrode (ISE) or ion chromatography. Both are routine.
• Cost: $15-25 standalone; included in most basic well-water panels.
• Frequency: at well purchase and every 3-5 years thereafter, more often if you're in a known hot zone.
• Sample handling: stable in plastic bottles; standard collection protocols.

Treatment

Three treatment methods work well for fluoride:

• Reverse osmosis — removes 90-95% of fluoride reliably. Best at point-of-use (kitchen tap). Whole-house RO impractical.
• Activated alumina — fluoride-specific adsorptive media; the standard residential POE option. Effective to roughly 1 mg/L residual. Cost: $1,500-3,000 installed; media replacement every 1-3 years depending on load.
• Anion exchange — works for fluoride but competes with sulfate and bicarbonate; less common than activated alumina for residential use.

What does not work: standard activated carbon (no), water softeners (no), boiling (concentrates fluoride). The same trio of useless treatments as for arsenic, with the same explanatory shape — fluoride is an inorganic anion that carbon doesn't adsorb and softeners don't capture.

Aquifers where this is a concern