Copper

WTR

0.3 mg/L California's health goal for copper

The enforceable action level is 1.3 mg/L, and it is judged across a sample of other people's taps.

Settled science

Stasis Home Research Every claim on this page is graded against peer-reviewed health research and linked to its primary source.Updated August 2026

Copper in tap water comes almost entirely from your own pipes and brass fixtures, not from the source or the plant. That is why it is measured at the tap. California's health goal is 0.3 mg/L; the federal action level of 1.3 mg/L describes a utility's sample of homes, not yours. Water that sat overnight carries the most, and reverse osmosis certified for copper is the strongest fix.

Copper: what is it, and where does it come from?

Copper is an essential nutrient and a plumbing material, and in drinking water it is almost always the plumbing material. Source water copper is low, around 0.01 mg/L in typical US surface water. What reaches your glass is copper the water dissolved out of your own home on the way to the tap: copper tubing, copper tanks, and the brass valves and fixtures that shed it even in houses plumbed in plastic. Three things drive how much. Corrosive water pulls more: slightly acidic or very soft water is the usual case, and WHO notes high-carbonate alkaline water can raise it too. Newer copper pipe sheds more than old, because the inside of the pipe gradually builds a less soluble mineral layer as it ages. That runs opposite to the usual lead story, where the worry is legacy lead-bearing materials in older homes, though neither metal follows a neat rule about pipe age. And stagnation concentrates it, so the water standing in the pipe overnight is the high-copper water and the first draw in the morning is the worst one. In a 1982 Seattle study reported by California's health hazard office, standing water in copper-pipe homes ran to a median of 0.76 mg/L against 0.35 mg/L for running water.

Copper: why does it matter?

We make no cancer claim about copper in either direction: EPA places it in the group it calls not classifiable for human carcinogenicity, which means the evidence is not sufficient to decide either way rather than a finding of no risk. It is a threshold toxicant with two distinct concerns. The first is acute and it is driven by the concentration in one glass rather than by what you drink across a day: a single glass above 3 mg/L is more likely to cause nausea than a whole liter carrying the same total copper spread through the day. Controlled human trials put the no-effect level at 2 to 4 mg/L and the lowest level where symptoms appear at 4 to 6, with symptoms fading after the first weeks of exposure. The second concern is narrower and it is the reason California's goal sits so much lower. A bottle-fed infant under six months, drinking formula reconstituted with tap water, gets a large share of its total copper from that water, and its ability to handle a surplus is not fully developed. California built its 0.3 mg/L goal on exactly that household and on gastrointestinal effects, not liver damage. What the evidence does not support, and what you will find claimed elsewhere, is that tap-water copper causes infant liver disease. The historical cases came from milk heated or stored in copper and brass vessels, in children who also carried an inherited susceptibility, and they largely disappeared when those vessels did.

Copper in drinking water is overwhelmingly generated by the home rather than delivered by the utility. WHO records that concentrations range from under 0.005 to over 30 mg/L, with the primary source most often being the corrosion of interior copper plumbing, and that levels in running or fully flushed water tend to be low while standing water is far more variable and can be substantially higher.WHO. Copper in Drinking-water (WHO/SDE/WSH/03.04/88, 2004). Guideline value 2 mg/L, no longer provisional; the single-glass concentration effect; premise-plumbing as the dominant source
In a two-month randomized double-blind community trial, 1,365 people in 441 families drank tap water at under 0.01, 2, 4 or 6 mg/L of copper. Gastrointestinal symptom reporting rose from 11.7% at near-zero copper to 15.3%, 18.3% and 19.7%. The authors report that at week one the difference from the near-zero group became significant in women at 4 mg/L and in men at 6 mg/L, and that significance faded as exposure continued.Araya et al., 2004 (Environ Health Perspect 112(10):1068-1073). Community-Based Randomized Double-Blind Study of Gastrointestinal Effects and Copper Exposure in Drinking Water, 1,365 individuals in 441 families
The acute effect tracks the concentration of a single drink rather than the daily total. WHO states that a single glass of tap water above 3 mg/L of copper is more likely to elicit nausea than a liter of water containing the same mass of copper ingested episodically through a day.WHO. Copper in Drinking-water (WHO/SDE/WSH/03.04/88, 2004). Guideline value 2 mg/L, no longer provisional; the single-glass concentration effect; premise-plumbing as the dominant source
California derived its 0.3 mg/L goal from a 426 µg/kg-day no-effect dose in a trial of 128 infants, a 50 percent allocation of total intake to drinking water, high-end infant water intake, and a threefold uncertainty factor covering individual variability and the study's limitations. Those limitations included that infants left the high-copper arm at three times the rate of the low-copper arm, which the study's own authors said could reflect unreported symptoms of intolerance.CA OEHHA. Public Health Goal for Copper in Drinking Water, February 2008 (300 ppb, superseding the 1997 value of 170 ppb; the infant derivation and its uncertainty factor; the Seattle standing-water data)
Infant liver disease from drinking-water copper is confined to a genetically susceptible minority at concentrations far above the health goal. The National Research Council places the associated drinking-water level at 3 mg/L and above, states that case reviews indicate not all infants are equally sensitive, and identifies first-draw water used to prepare infant formula as the specific route of concern.US National Research Council (Committee on Copper in Drinking Water), Copper in Drinking Water, National Academies Press, 2000. The 3 mg/L susceptible-infant level; unequal infant sensitivity; first-draw water in formula preparation; Wilson's disease outside the goal process
Flushing is not a reliable copper control. A peer-reviewed review of drinking water in schools and childcare centers found that flushing outlets in copper-piped buildings did not consistently reduce copper and in some cases raised it, with concentrations rebounding above the action level in under thirty minutes.Montagnino E, Lytle DA, Rose J, Cwiertny D, Whelton AJ, 2022 (AWWA Water Science 4(2):e1270). School and childcare center drinking water: copper chemistry, health effects, occurrence, and remediation
Reverse osmosis has a documented field result at household scale, though only one. EPA reports that four-stage point-of-use reverse osmosis units installed at a 16-unit Montana trailer park reduced copper from 3.25 mg/L to 0.22 mg/L, a 93 percent reduction. That is one unusually corrosive site reported in 2006, not a performance specification for the technology.US EPA 815-R-06-010. Point-of-Use or Point-of-Entry Treatment Options for Small Drinking Water Systems, 2006, section 7.2.2 (one documented installation: 3.25 mg/L to 0.22 mg/L)
A certification for one heavy metal does not cover another. Under NSF/ANSI 53, the standard behind carbon filters and pour-through pitchers, 1,016 products were listed for lead reduction on 2026-08-11 and exactly one of them also carried a copper claim; of the 139 listed with a copper claim, every one was a pour-through pitcher or dispenser. Reverse osmosis is certified under NSF/ANSI 58, where the copper and lead listings returned the same 144 products.NSF. Copper in Drinking Water, and the NSF certified drinking water treatment unit listings for NSF/ANSI 53 and 58 (queried 2026-08-11; certification is a per-contaminant claim, and pass/fail rather than a percentage)

Copper: what do we grade it against?

Copper is held to two different numbers, and they do not agree. The health-based level for Copper (mg/L) is 0.3 (California public health goal, set on a formula-fed infant), from CA OEHHA Public Health Goal for copper, February 2008. The enforceable legal limit is 1.3. The table below lists every level on this page next to the source that set it.

Where the health research draws the line, versus the legal limit.
Contaminant	Health-based level	Legal limit	Source (health-based)
Copper (mg/L)	0.3 California public health goal, set on a formula-fed infant	1.3 EPA and California action level, 90th percentile across sampled taps	CA OEHHA Public Health Goal for copper, February 2008
Copper, WHO guideline (mg/L)	2 protective against acute gut effects in adults with normal copper handling	none set	WHO, Copper in Drinking-water, 2004

One thing to keep straight about the 1.3 mg/L line: it is not a limit for your house. A system exceeds it when the 90th-percentile result across its compliance samples comes out greater than 1.3 mg/L, so one home above 1.3 does not by itself establish an exceedance, and a utility in compliance tells you nothing about your own plumbing. That is the whole reason copper is worth testing at your tap rather than reading off a report. Note also that EPA's own health goal for copper is set at the same 1.3, so unlike most contaminants on this site the federal health number and the federal action number are one number; the 0.3 in the health column above is California's, and it is the more protective one. One number deliberately not in this table: there is a 1.0 mg/L secondary standard for copper, which is about taste and blue-green staining rather than health. EPA's version is non-enforceable federal guidance; California adopts 1.0 mg/L as a secondary maximum contaminant level in its own regulations. Either way it is an aesthetic number, which is why it has no business in a health column.

What helps against Copper?

4 approaches below reduce Copper in a home, ordered from the most direct fix to the smallest change in habit. The most direct is Reverse osmosis at the kitchen tap. Each entry also names what it cannot do, because no single fix covers every exposure route.

Direct fixes

Reverse osmosis at the kitchen tap
An under-sink reverse osmosis system rejects dissolved copper at the membrane, and it is typically the same unit a household would already be considering for PFAS, chromium-6 and other metals, so copper can come along with it rather than needing its own device. It is also the one form factor where a single certified unit commonly covers both copper and lead.
Copper is an optional claim under the reverse osmosis standard, not an automatic one, so the listing has to name copper specifically, and every other contaminant you care about has to be named too. The certification proves the unit brings copper below the 1.3 mg/L action level, not below California's 0.3 mg/L goal, so retest the treated water if 0.3 is your target. Treat any published percentage with suspicion unless it is that exact product's own certified test result.
A filter specifically certified for copper reduction
A point-of-use filter certified for copper is challenged with 3.0 mg/L and must bring the treated water to 1.3 mg/L or below, tested at two different pH levels.
Rarer than people assume, and almost never the filter you already own. A filter certified for lead is very unlikely to carry a copper claim, and under NSF/ANSI 53, the standard those filters are certified to, every copper-certified product we found on 2026-08-11 was a pour-through pitcher or dispenser rather than an under-sink or faucet-mount unit. Reverse osmosis is the exception, certified under NSF/ANSI 58. A plain carbon block with no copper claim should not be relied on.

Bigger retrofits

Corrosion control, or replacing the pipe
These are the two levers that act on the pipe rather than on the water at one tap. Raising pH and dosing orthophosphate passivate the inside of the plumbing so it releases less copper, while replacing the pipe is what actually removes the source. Where the water is genuinely aggressive, WHO notes copper tubing may not be the appropriate material at all.
On a public system the water chemistry is mainly the utility's lever rather than yours, though a private well or a whole building can be treated with professional design and monitoring. Repiping is a real project and belongs in a conversation about the whole plumbing system, not as an answer to one test result.

Free and behavioral

Running the tap before you draw drinking water
Flushing clears the standing slug that picked up copper overnight, and it is free. Use cold water only, and never the hot tap, for drinking, cooking or mixing formula.
A stopgap, not a fix, and it is not guaranteed to get you under the health goal. In the 1982 Seattle data California cites, running water in copper-pipe homes still had a median of 0.35 mg/L, above the 0.3 mg/L health goal. A peer-reviewed review of drinking water in schools and childcare centers found flushing did not consistently reduce copper and sometimes raised it, with rebound in under half an hour.

Copper is the cleanest example of why we test your tap instead of reading you your district's report. Two houses on the same street, on identical water, can be an order of magnitude apart because one has copper pipe and a long overnight stagnation and the other does not. For most households a small detection is a plumbing signal worth knowing rather than a health alarm, and we will say that plainly rather than inflate it. The household California had in mind when it set the 0.3 mg/L goal is specific: a formula-fed infant under six months. If that is your home and your copper is above 0.3 mg/L, that is worth acting on, and there is something to do this week before any purchase: mix formula with cold water only, never the hot tap, and run the tap before you draw it. The most durable point-of-use option is reverse osmosis at the kitchen tap with copper named on its certification. Note what that certification proves: it demonstrates the unit brings copper below the 1.3 mg/L action level, not below the 0.3 mg/L goal, so if 0.3 is your target, test the treated water.

Common questions about Copper

Where does copper in tap water come from?

Your own plumbing, in almost every case. Source water carries very little copper, typically around 0.01 mg/L. What you drink is copper the water dissolved out of copper pipe, copper tanks and brass fixtures on its way to the tap. That happens fastest when the water is slightly acidic or very soft, when the copper pipe is relatively new, and when the water has been sitting still.

Is copper in drinking water dangerous?

At the levels most homes see, no. Copper is a threshold toxicant rather than a no-safe-level contaminant, and EPA places it in the group it calls not classifiable for human carcinogenicity, so we make no cancer claim about it in either direction. The gut effects, nausea and stomach upset, show up in controlled studies somewhere between 2 and 6 mg/L, well above what a typical tap produces. Two households should read the number more carefully. A formula-fed infant under six months is the exposure California modeled when it set its 0.3 mg/L goal, though California is clear the goal is meant to be protective more broadly and not only for that home. And for anyone with Wilson's disease the levels that matter clinically sit far below any drinking-water regulation, which is a conversation for their physician rather than a number we can give you.

My utility says it meets the copper action level. Am I fine?

It does not tell you much about your house. The action level turns on the 90th-percentile result across the taps the utility sampled, homes chosen mainly for lead risk. Copper is generated by each home's own pipes, so a compliant utility and a high reading at your kitchen sink are entirely compatible. This is one of the few contaminants where testing your own tap is the only way to know.

Does copper in water cause liver damage in babies?

Not at the levels found in ordinary household plumbing, and the claim is worth correcting because it circulates. The historical infant liver diseases linked to copper involved both an inherited susceptibility and very high intake, and the copper came from milk heated or stored in copper and brass vessels rather than from pipes. Those two conditions, Indian childhood cirrhosis and Tyrolean infantile cirrhosis, largely disappeared once those vessels went out of use. The National Research Council places the drinking-water level associated with liver toxicity in susceptible infants at 3 mg/L and above. The largest study of the question, run in Berlin and reported in California's own review, sampled water from 2,944 households that had copper plumbing and an infant, identified 541 infants who met the follow-up criteria of water at or above 0.8 mg/L plus a minimum daily intake, examined them, and ran blood serum analysis on 183. None of the examined infants showed any symptom of liver damage, and with one exception no correlation appeared between liver enzymes and estimated copper intake from tap water.

Does a water filter remove copper?

Only if it is specifically certified for copper, and most are not. Certification works contaminant by contaminant, and a filter certified for lead is very unlikely to also carry a copper claim. When we checked the certified-product listings on 2026-08-11, of more than a thousand products certified for lead under NSF/ANSI 53 exactly one also claimed copper, and every copper-certified product under that standard was a pour-through pitcher or dispenser rather than an under-sink unit. Reverse osmosis is the practical answer for a household that needs both, with copper named on the listing.

Should I just run the tap first?

It helps and it costs nothing, so do it, using cold water only and never the hot tap. But do not treat it as the fix. In the 1982 Seattle data California relies on, flushed water in copper-pipe homes still had a median above the 0.3 mg/L health goal, and a peer-reviewed review of drinking water in schools and childcare centers found flushing did not consistently lower copper and sometimes raised it, with levels climbing back within half an hour. It is a useful interim measure, especially before mixing formula, not a solution.

What if someone in my home has Wilson's disease?

Talk to their physician about the water, and treat that conversation as the answer rather than anything you read here. No drinking-water standard is designed for Wilson's disease. California explicitly left it out of the basis for its health goal on the grounds that affected people are under medical care, and the National Research Council notes that copper for Wilson's patients is not managed through the drinking-water goal process at all. The levels clinicians care about are far below any regulatory number.