Manganese in well water: black stains, the health numbers agencies disagree on, and what fixes it

Most people find manganese in their well the same way: black streaks in the toilet bowl, gray laundry, a bitter taste in the coffee. The harder part: the level where manganese stains things and the level where health agencies start to worry are different numbers, and the agencies don't agree on the second one. For a baby on formula, Minnesota's line is a third of EPA's. Below: the numbers side by side with who said what and when, then testing, treatment, and why water can turn black right after you add a softener or chlorinate.

What manganese is and why your well has it

Manganese is a metal in ordinary rock and soil. EPA's 2004 health advisory calls it "one of the most abundant metals on the earth's surface" and "a component of over 100 minerals." It is also a nutrient. The Minnesota Department of Health (MDH): "Your body needs some manganese to stay healthy, but too much can be harmful."

The question for a well is whether the groundwater can dissolve it, and that comes down to oxygen and pH. In its national study of private wells (Circular 1332), the U.S. Geological Survey (USGS) found that "relatively high concentrations of manganese coincided with low dissolved oxygen concentrations and low pH, because manganese-containing minerals are more soluble in water under these geochemical conditions." EPA's advisory says it more briefly: "Anaerobic groundwater often contains elevated levels of dissolved manganese."

The same chemistry explains why iron usually comes along. Michigan's environmental agency (EGLE) says manganese "is commonly found with iron as mineral oxide coatings on the surface of soil and rock grains. When groundwater contacts these coatings, the oxides are dissolved and may be transported to your well."

Underground, without oxygen, the manganese stays dissolved and invisible. Penn State Extension describes what happens next: the water "may appear colorless initially but orange-brown (iron) or black (manganese) stains or particles quickly appear as the water is exposed to oxygen." Penn State says this dissolved, "reduced" form is "most common in groundwater with a pH less than 7.0." That drives most treatment choices and black-water surprises.

Stains and taste: the 0.05 mg/L guideline

EPA's secondary standards page (updated May 26, 2026) sets a secondary maximum contaminant level (SMCL) for manganese of 0.05 mg/L (50 µg/L, or 50 parts per billion). Under "Noticeable Effects above the Secondary MCL," it lists "black to brown color; black staining; bitter metallic taste."

EPA does not enforce secondary standards, and says these contaminants "are not considered to present a risk to human health at the SMCL." On a private well nothing federal is binding anyway: EPA's private wells page says private well water is not regulated "by the federal government under the safe drinking water act nor by most state governments."

Staining can start lower. EPA's 2004 advisory, citing older studies, says "at concentrations as low as 0.02 mg/L, manganese can form coatings on water pipes that may later slough off as a black precipitate." Vermont's health department says levels between 0.050 and 0.300 mg/L may bring a metallic taste, a bad smell, deposits and "brown or black staining," adding: "None of these pose a health risk."

Stains aren't a reliable signal either way. In a 2017 study of private wells in Dakota County, Minnesota (the WIISE study, by MDH and the county), higher manganese went with only "a modest increase in aesthetic concerns and mitigative actions." The authors concluded that "due to the lack of a consistent, reliable geochemical indicator for elevated manganese, widespread well sampling is required." Clean fixtures prove nothing; only a test gives you a number.

Health effects: what EPA, ATSDR and the states say

The federal health number is in EPA's Drinking Water Health Advisory for Manganese (EPA-822-R-04-003, January 2004): a lifetime health advisory of 0.3 mg/L, which EPA says "will protect against concerns of potential neurological effects," and a one-day and ten-day advisory of 1 mg/L "for acute exposure." For young babies EPA advises "that for infants younger than 6 months, the lifetime HA of 0.3 mg/L be used even for an acute exposure of 10 days," because of differences between breast milk and formula and "the possibility of a higher absorption and lower excretion in young infants."

An advisory is guidance, not a rule: EPA's health advisories page (updated March 30, 2026) says advisories "are not to be construed as legally enforceable federal standards." The 2004 values are still the ones in EPA's 2018 advisory tables.

The 2004 document is cautious about the evidence. It names the nervous system as "the primary target organ," but notes that many human reports "are from inhalation exposures in occupational settings" and that "there are few data for the association between oral exposure to manganese and toxic effects." The groups it calls "possibly sensitive": "the very young (who may absorb more and excrete less), the elderly, and those with liver disease."

The research on children, as agencies summarize it

Newer studies focus on children. The hedges are the agencies' own:

• ATSDR (ToxFAQs, last reviewed March 20, 2014): "Studies in children have suggested that extremely high levels of manganese exposure may produce undesirable effects on brain development... We do not know for certain that these changes were caused by manganese alone."
• Maine CDC: "Some recent epidemiological studies have suggested an association between exposure to manganese in drinking water and intellectual impairment and poorer neurobehavioral function in young children."
• EPA (January 2025): information developed for the World Health Organization's 2021 guidelines "indicates exposure to elevated levels of manganese in drinking water contributes to increased risk for adverse neurological effects, such as behavioral and sensorimotor effects."

Sources don't all agree. Penn State Extension (updated May 29, 2025) says "iron and manganese are not health concerns in drinking water," though it also describes EPA's 0.3 mg/L advisory and the "neurological health effects" behind it. State agencies in Minnesota, Wisconsin, New Hampshire and Vermont publish health-based values or action levels, and Maine CDC cites EPA's.

Bathing isn't the concern: MDH says such water "can be harmful for your health, but taking a bath or a shower in it is not."

Is a federal limit coming?

As of November 1, 2026, there is no federal maximum contaminant level for manganese, and EPA has not decided to develop one. From EPA's own pages:

• July 18, 2003: EPA's first regulatory determinations found "no regulatory action is appropriate or necessary" for manganese and eight other contaminants, and EPA issued health advisories instead.
• 2016 and 2022: manganese returned to the Contaminant Candidate List (CCL 4) "based on nominations" and is on the final CCL 5 (November 14, 2022).
• 2025 and 2026: EPA's January 15, 2025 preliminary determinations (90 FR 3830) made "no determination" for manganese, which EPA "continues to evaluate." The March 2026 final round covered nine other contaminants; EPA's support document says it "is making no determination about the remaining five contaminants," manganese among them.
• Now: manganese is on the draft CCL 6 list (announced April 2, 2026). A Federal Register search on November 1, 2026 found no EPA drinking-water action on manganese after that notice.

Listing means EPA may regulate it someday, not that it will; a federal rule would cover public systems, not private wells.

Infants and formula: why the cutoffs differ

Two reasons babies get the most attention. Maine CDC says adults normally absorb "less than 5 percent" of the manganese they swallow, while infants "can absorb as much as 20 percent" and excrete less. And formula already contains manganese: EPA's advisory says "infant formula typically contains a much higher concentration of manganese than human or cows' milk," so powdered formula mixed with well water adds to it.

Each agency's guidance in its own terms. Note the age cutoffs:

The "per year" wording is how New Hampshire and Maine apply EPA's advisory; EPA's 2004 document says only "even for an acute exposure of 10 days." Minnesota's 100 µg/L comes from its own risk assessment, based on neurological effects in a rat study; for children over one and adults, MDH "continues to support" EPA's 300 µg/L (see certified labs in Minnesota).

For comparison only, outside the U.S.: Health Canada's 2019 guideline sets a maximum acceptable concentration of 0.12 mg/L for total manganese, "based on infants, the most sensitive population," and an aesthetic objective of 0.02 mg/L. That is Canadian guidance, not U.S.

Maine's first step: "switch to bottled water or water that is low in manganese to make formula." On breastfeeding, Maine says "there is no correlation between manganese levels in water and manganese levels in breast milk" and that a healthy mother should continue. Questions about your own child go to your health care provider.

How common is high manganese in wells?

In USGS's national sample of about 2,100 private wells (1991 to 2004, water collected before any home treatment), manganese and iron "were each detected in about half of the wells." Manganese was above EPA's 50 µg/L secondary standard in 21% of wells, and above the USGS health-based screening level of 300 µg/L in 5.2% of the 2,159 wells sampled for it.

USGS found elevated manganese "most often in the glacial aquifers, in sandstone aquifers in Pennsylvania and West Virginia, and in some coastal-plain aquifers in Louisiana and Texas." Its regional study of the glacial aquifer system (SIR 2009-5006), which underlies most of New England and extends through the Midwest, found that of 556 drinking-water wells, "9.9 percent (55) exceeded the manganese Lifetime Health Advisory."

State data shows how local it gets. MDH says groundwater in southeastern Minnesota "tends to have low levels of manganese (below 50 µg/L)," the southwest "tends to have higher levels—some over 1,000 µg/L," and "there are no clear patterns in the other parts of the state." In the WIISE study, 194 of 274 untreated outdoor faucets at private wells in Inver Grove Heights (71%) were above 100 µg/L, and 56% were above 300 µg/L. Maine says elevated manganese "can occur in private wells nearly everywhere in Maine." None of that tells you what your well holds; a neighbor's result is a reason to test, not a substitute.

Testing: what to order and where to sample

Manganese often isn't in a basic bacteria-and-nitrate test, so ask for it by name. MDH says to test for it "at least once," alongside arsenic and lead. Vermont recommends "testing your well or spring for manganese every five years." Maine CDC and New Hampshire DES both recommend a standard test that includes manganese every 3 to 5 years; Maine adds "when water changes in smell, taste or color." Wisconsin DHS says to test if your water "is brown or rust color, stains the faucets, sinks, or laundry, or has an off-taste or odor," and that households with pregnant people and young children "may be eligible for free testing through their local health department" (see certified labs in Wisconsin).

What to ask the lab for

• Total manganese. Health Canada's guideline is written for total manganese. New Hampshire says total versus filtered results show how much is "dissolved or particulate," which matters for choosing treatment.
• Iron, hardness, pH and alkalinity. Michigan EGLE: "It is recommended that the total iron, manganese, hardness, alkalinity and pH of the water be tested before purchasing equipment." Every treatment choice below depends on them.
• Arsenic and the basics. Wisconsin's fact sheet: "You should also test for arsenic, bacteria, and nitrate regularly." USGS says arsenic release from iron oxides "is favored by low dissolved oxygen concentrations and high pH," and the WIISE study found manganese correlated with arsenic and iron. See the arsenic explainer, and the lead explainer for its special sampling.

Where to take the sample

WIISE sampled untreated outdoor spigots, then the inside drinking tap at homes above 100 µg/L. Most of those homes already used some treatment, yet "of the 109 inside tap samples submitted, 36% still exceeded" 100 µg/L. A raw-water sample, taken before any softener or filter, shows what the well delivers and what treatment must handle. A sample from the tap you drink from shows what you actually drink. With treatment in place and a baby on formula, get both.

The manganese lab finder lists state-certified labs, the which-tests tool helps you build a panel, and the guide to testing well water covers sampling step by step.

Reading your result: a decision table

Units: 1 mg/L is 1,000 µg/L, and µg/L is parts per billion (ppb), so 0.42 mg/L, 420 µg/L and 420 ppb are the same. None of these values is legally binding on a private well, and what's right for your household, especially with an infant, is a question for your doctor or health department. You can also enter your number in the results checker.

Treatment: matching the method to your water

Penn State says removal methods depend "on both the form and concentration of the metals." To stop staining, Penn State and New Hampshire both recommend treating all the water entering the house.

Minnesota's field data is the strongest evidence for softeners in real homes. In WIISE, "all softened samples were below both health-based guidance values" (100 and 300 µg/L), with "no indication of reduced softener effectiveness across the entire range of manganese, up to 1000 µg/L." The report also notes earlier warnings that iron above 5 mg/L and manganese above 500 µg/L "can foul the softener resin bed," says "more research on softeners is needed," and found that "carbon filters, iron filters, and sediment filters did not consistently reduce manganese levels." Minnesota points to softeners; New Hampshire prefers oxidation-filtration.

Whatever you install, test afterward. Vermont: "Re-test your drinking water for manganese after any treatment system is installed to make sure levels are below the advisory level." Look for independent certification, not marketing claims; Wisconsin DNR points owners to the state's approved-device list, and Vermont to NSF-certified systems.

Why water turns black after a softener or chlorine

Possible causes, the first four from the sources; a test or a treatment professional can sort out which applies.

• Manganese oxidized before the softener. A softener handles dissolved manganese. Penn State says "it is critical that the raw water not come in contact with any oxidizing agents like air or chlorine before entering the softener," because oxidized forms "will foul the softener resin."
• Chlorine turned dissolved manganese into solids. Penn State says an oxidizing chemical, most commonly chlorine, converts "dissolved iron and manganese into the solid, oxidized forms," and Nebraska Extension says chlorine "reacts with certain components in the water including iron, manganese, sulfur." Water that was clear can turn black once those solids form.
• Old deposits came loose. EPA's advisory notes manganese coatings on pipes "may later slough off as a black precipitate."
• Manganese bacteria. Nebraska Extension says iron and manganese bacteria "do not pose a health risk but they are often the cause of blackish or reddish slime." Shock chlorination is "an effective but often short term method," and "repeated shock chlorination treatments can cause corrosion or damage to well components."
• The water isn't being softened (our inference). WIISE didn't study black water, but inside taps that still tested hard (250 or 425 mg/L) had median manganese of 399 and 241 µg/L; at 0 or 50, medians were below detection.

Before chlorinating a well that has treatment equipment, Nebraska Extension says to check with the equipment manufacturers and generally to remove carbon and RO filters while chlorine is in the system.

Worked example: one report, step by step

Illustrative numbers, not a real test. A family with a three-month-old on formula gets this raw-water result: total manganese 0.42 mg/L, total iron 2.1 mg/L, pH 6.9, hardness 220 mg/L as calcium carbonate. The water comes out clear and stains the toilet black within a day.

1. Convert. 0.42 mg/L is 420 µg/L: the "300 to below 1,000 µg/L" row of the decision table.
2. Health lines. 420 µg/L is above EPA's 0.3 mg/L (lifetime, and for infants under six months even for 10 days), Minnesota's 100 and 300 µg/L, Wisconsin's 300 µg/L action level, and Vermont's and New Hampshire's 0.3 mg/L. It is below EPA's 1 mg/L short-term advisory.
3. The baby first. Minnesota, Wisconsin and Maine all advise another water source for formula at this level. Maine: "switch to bottled water or water that is low in manganese to make formula." Boiling won't help. The parents should ask their pediatrician about the baby.
4. Confirm. Wisconsin says to "collect a second sample to verify the results." Keep using the other water meanwhile.
5. Read the chemistry. Clear water that blackens on standing fits Penn State's description of dissolved manganese. This water meets Penn State's softener conditions (pH 6.9 is above 6.7, hardness 220 is within 50 to 350 mg/L, iron is under 5 mg/L) and New Hampshire's 150 mg/L hardness threshold. Birm fits poorly: Penn State wants pH 7.5 for manganese. Combined iron and manganese, 2.52 mg/L, is just under Penn State's 3 to 10 mg/L greensand range. New Hampshire would steer this owner toward oxidation-filtration with a strong pre-oxidant.
6. Verify the treated tap. Whatever they choose, the number that matters is a new sample from the tap they use for formula. Under Minnesota's infant guidance, the target is 100 µg/L or less. Until a treated-tap result shows it, formula water comes from somewhere else.
7. Keep testing after any equipment change, if stains return, and on your state's schedule.

Ready to test? Find a certified lab with the manganese lab finder. If you're ordering a full metals panel, read the arsenic and lead pages first so you collect the right samples in one trip.

Questions people ask