Radon in well water: how much it adds to your air, and when to treat it

Radon in well water is mostly an air problem. You don't get most of the risk by drinking the water; you get it by breathing the gas the water lets go of in the shower, the dishwasher and the washing machine. The U.S. Environmental Protection Agency (EPA) compares it to a carbonated soda: the gas is dissolved in the water and comes out when the water is used. That one fact decides almost everything about radon in a well: which test to run first, which number on your report matters, and why a filter on the kitchen tap doesn't solve it.

What radon is, and why some wells are full of it

Radon is a radioactive gas with no color, smell or taste. EPA says it "comes from the natural decay of uranium and radium found in nearly all rocks and soils," and lists its half-life as 3.8 days. EPA's archived drinking-water page says radon gas "can also dissolve and accumulate in water from underground sources (called ground water), such as wells."

Lakes and rivers are a different story. EPA says radon "is not a concern" in surface water, "because the radon is released into the air before it ever arrives at your tap."

Geology does most of the sorting. The U.S. Geological Survey (USGS) sampled domestic wells from 1991 to 2004 for Circular 1332 and found radon "in samples from nearly all wells." Its occurrence "is controlled, in large part, by the presence of uranium-bearing rocks." Concentrations were highest "in crystalline-rock aquifers located in the Northeast, in the central and southern Appalachians, and in central Colorado," where about 30 percent of sampled wells were above 4,000 pCi/L. State programs describe the same pattern closer to home:

• New Hampshire: radon "is commonly found in bedrock and in water from bedrock (drilled) wells," while "dug wells and point wells tend to have minimal to no radon" (NH Department of Environmental Services).
• Wisconsin: the highest levels are "in groundwater flowing through granite or granitic sand and gravel formations commonly found in northcentral and northwestern Wisconsin," and "every well in Wisconsin has some level of dissolved radon in it" (Wisconsin DNR).
• Vermont: radon gets into drinking water "if your well is drilled into or near natural rock formations that contain uranium" (Vermont Department of Health).

New Hampshire also advises testing bedrock wells for uranium and gross alpha, because "a well that has high levels of radon is more likely to have high levels of uranium and/or gross alpha." See uranium in well water.

Air vs. water: the 10,000-to-1 rule

EPA's 2012 Report to Congress on radon in drinking water says the National Academy of Sciences (NAS) estimated a "transfer coefficient" from water to indoor air of about 10,000 to 1: "given typical water use patterns and indoor physical configurations, on average, 10,000 picocuries of radon per liter of tap water will contribute approximately one picocurie per liter to the concentration of radon in indoor air." The same report spells out what follows: "4 pCi/L of radon in indoor air corresponds to about 40,000 pCi/L in drinking water."

State programs use the same rule of thumb. Maine CDC: "for every 10,000 pCi/l of radon in your water, it will add 1 pCi/l to your home's air." New Hampshire calls it "a general rule of thumb." Because EPA ties the ratio to "typical water use patterns and indoor physical configurations," treat it as an average, not a measurement of your house. Connecticut adds that radon in the air "will increase for a short period of time" when you use the dishwasher, washing machine, shower and bath.

Now compare that with EPA's advice for air. EPA says to fix your home if indoor air is 4 pCi/L or higher, and to consider fixing between 2 and 4 pCi/L. By the 10,000-to-1 rule, it takes about 40,000 pCi/L in the water to produce 4 pCi/L in the air on its own. EPA's archived drinking-water page estimates that "only about 1-2 percent of radon in the air comes from drinking water." Individual homes can look very different: Maine CDC says "many Maine homes have high air radon levels due only the radon which escapes from the water."

The health risk: mostly lungs, a little stomach

EPA's health risk page says radon "is the number one cause of lung cancer among non-smokers, according to EPA estimates," and the second leading cause overall. EPA puts radon-related lung cancer deaths at about 21,000 a year in the U.S., about 2,900 of them among people who never smoked. Smoking raises the risk sharply: EPA's tables say that if 1,000 smokers were exposed to 4 pCi/L over a lifetime, about 62 could get lung cancer, against about 7 of 1,000 people who never smoked.

For radon that starts in water, EPA's estimate comes from a second 1999 NAS report. In its 1999 fact sheet on the proposed rule, EPA wrote that radon in drinking water "causes about 168 cancer deaths per year, 89 percent from lung cancer caused by breathing in radon released from water, and 11 percent from stomach cancer caused by drinking radon-containing water." EPA says the ingestion risk "is smaller than the risk of developing lung cancer from radon released to air from tap water." Maine CDC, explaining why its 4,000 pCi/L guideline adds only about 0.4 pCi/L to indoor air, puts it more strongly: "According to studies, ingesting water with radon in it does not present a significant health hazard."

Either way, most of the risk from radon in your well comes through the air, so treating only the water you drink misses most of it. For questions about your own exposure, especially if anyone in the house smokes, ask your doctor or your state radon program.

No federal limit: what EPA proposed in 1999

EPA's archived page is plain about it: "There is currently no federally-enforced drinking water standard for radon," and "EPA does not regulate private wells." On November 2, 1999, EPA proposed a radon rule for community water systems that use groundwater or a mix of ground and surface water. EPA has not finalized it. The archived page says "EPA must promulgate a final regulation before a federal radon regulation will be enforced," and as of November 1, 2026, EPA's table of drinking water regulations (updated August 31, 2026) has no radon entry. EPA's 2018 standards and health advisories table still marks radon "P," for proposed.

The 4,000 figure wasn't arbitrary. EPA's 2012 report explains that 4,000 pCi/L in water corresponds, by the NAS ratio, to 0.4 pCi/L in air, the average radon level in outdoor air. The two-number design comes from the 1996 Safe Drinking Water Act amendments, which, in EPA's words, provide "for a multimedia approach" to radon in drinking water and radon in indoor air from soil. None of these numbers ever applied to private wells.

State guidance you can open

Several state programs publish their own numbers for radon in private well water. They are guidelines and advisory levels, not legal limits, and they don't agree. Each row below comes from that state's own page or fact sheet.

The triggers run from 2,000 pCi/L (New Hampshire, and only with high air) to 10,000, so "is my number safe?" depends on whose guidance you read. And most of them tie the water number back to air: New Hampshire's advice depends on the air result, Massachusetts uses its guideline to trigger an air test, and Maine tells you to weigh "your risks from all radon sources (soil gas and water)." Summaries flatten this. USGS's 2009 circular listed New Hampshire simply as 2,000 pCi/L. Use the current page from your own state.

EPA radon zones: what the map tells you, and what it doesn't

Each state page on this site shows a chart of counties by EPA radon zone. EPA built the Map of Radon Zones in 1993 "to identify areas of the U.S. with the potential for elevated indoor radon levels," using indoor measurements, geology, aerial radioactivity, soil data and foundation types. The zones, as EPA's map defines them:

• Zone 1: highest potential; average indoor radon levels may be greater than 4 pCi/L.
• Zone 2: moderate potential; average indoor levels may be between 2 and 4 pCi/L.
• Zone 3: low potential; average indoor levels may be less than 2 pCi/L.

Two limits matter. The map is about indoor air, not water. And EPA says the map "should not be used to determine if individual homes need to be tested," adding that "homes with elevated levels of radon have been found in all three zones." Maine CDC adds a point that matters for wells: "the presence of soil gas radon does not indicate or predict the presence of well water radon (nor does well water radon indicate or predict soil gas radon)." Vermont's health department puts the conclusion plainly: "Testing is the only way to know if radon is in your drinking water."

Here is how EPA's county list (EPA 402/A-16/001) breaks down for the 10 states this site covers. These are our counts of EPA's spreadsheet.

Vermont is the cautionary case. None of its 14 counties is in Zone 1, yet Vermont's health department sets a 4,000 pCi/L advisory level for well water and recommends testing for radon in water every five years. Use a zone as a reason to test sooner, never to skip testing.

Test the air first, then the water

EPA's Citizen's Guide to Radon is direct about the order. Its myths page answers "Everyone should test their water for radon" with: "Although radon gets into some homes through water, it is important to first test the air in the home for radon." Its section on water adds: "If you've tested the air in your home and found a radon problem, and your water comes from a well, have your water tested." Wisconsin DNR says to test household air first, and Vermont says to test the air before deciding how to reduce radon levels. Maine and Connecticut go further and recommend a water test for every home on a well; Connecticut notes "it is possible to have elevated radon levels in your water even if the radon level in your indoor air is low."

An air test measures what you breathe from both sources at once. A water test tells you how much of it could be coming from the well. You need both to choose a fix, and Vermont publishes a radon contribution calculator for that split.

How to take a radon-in-water sample

Radon "easily escapes water when it is agitated or aspirated" (Maine CDC), so a radon sample has to be caught gently and sealed without a bubble, in small glass vials from the lab. The steps below follow the Vermont Department of Health Laboratory's Kit RC instructions (version 7.0); Maine's instructions for public-system samples are similar.

1. Use the kit's containers: Vermont's kit has two 40 mL glass vials. Don't touch the inside of the cap or the mouth of the vial.
2. Take the aerator or strainer off the faucet if you can.
3. Run the cold water at full flow for 3 to 5 minutes. Maine's instructions give the reason: "to make sure water has not been sitting for a long time in pipes or tanks and is fresh from the well."
4. Turn the flow down to low. Maine's instructions say a 1/8-inch stream.
5. Fill each vial until it overflows, then cap it tightly. Fill both vials from the same tap.
6. Turn each vial over and tap it gently. Vermont: "If air bubbles of any size are present, dump out the vials and repeat sample collection." Maine: "Samples with bubbles cannot be analyzed."
7. Write down the date and time you filled the vials, and get them to the lab fast. Vermont requires delivery within 48 hours of collection; Maine's public-system instructions give a maximum holding time of 4 days. Radon decays as it waits: with a 3.8-day half-life, a vial that sits for a week has lost more than half its radon before anyone measures it.

Plan for more than one sample. Vermont: "Because radon concentrations in well water can vary throughout the year, it can be beneficial to test more than once at different times of the year." New Hampshire recommends "at least two radon tests (at least one month apart when possible)" before any treatment decision, and Massachusetts suggests at least two samples "taken a month or two apart, if possible" for radionuclides generally. Connecticut's action level uses the average of two samples drawn at the same time.

In Maine, who takes the sample matters. Maine CDC says the owner or occupant may test, but "if the home is for sale, an independent registered radon tester is required," and it is illegal for a realtor or home buyer to run the radon test. Other states have their own rules, so check before a sale. Use the radon lab finder to find state-certified labs, and see the guide to testing well water and the which-tests tool if you're ordering uranium or gross alpha at the same time.

Reading your result: a decision table

Read your water number against your air number; a water result of 4,000 and an air result of 4 measure different things: radon per liter of water vs. per liter of air. The "adds to air" column applies EPA's NAS-based average of 10,000 to 1; your own house may differ. The state guidance column quotes each state separately. You can also enter results in the results checker.

Treatment: aeration, carbon, and why a tap filter won't do

Treatment has to catch radon before the water reaches any faucet. EPA's Citizen's Guide says point-of-use devices at the tap "only treat a small portion of the water you use and are not effective in reducing the risk from breathing radon released into the air from all water used in the home." Wisconsin DNR is blunter: "Small carbon filters attached to kitchen faucets or placed under a sink are inadequate to treat your water for radon," and bottled water "will reduce your risk from radon ingestion but will not eliminate the problem of radon gas escaping from your well water into your household air." New Hampshire: "Radon is volatile and must be treated at point-of-entry or whole-house."

On carbon, the agencies point the same way and differ only in how cautious they are. EPA's Consumer's Guide to Radon Reduction (the removal figures in the table come from it) suggests carbon for "moderate levels, around 5,000 pCi/L or less in water"; Connecticut, Maine and New Hampshire draw their line at 10,000; Vermont "discourages the use of GAC systems to remove radon because the radon collected on the filter could create a radiation hazard," and asks for NSF/ANSI Standard 53 certification if you install one anyway. Carbon is for lower levels with careful servicing, aeration for higher ones. Connecticut recommends a nationally certified radon mitigation professional for the installation, and Wisconsin requires any aeration or GAC system to be approved by the state plumbing program.

Water treatment doesn't fix radon coming through the foundation; Massachusetts notes soil-gas radon "may require a different treatment system than that which would be used for water as a sole source of radon." Wisconsin calls a replacement well "a last resort," since it "will not necessarily yield water with lower radon levels." Retest after any system goes in (Vermont), and Maine CDC recommends re-testing every 2 years after a mitigation system is installed.

Worked examples

These are illustrative numbers, not real test results. Each follows the decision table above.

House A, Maine, drilled well. Winter air test in the basement: 6.0 pCi/L. Two water samples average 12,000 pCi/L.

• By the NAS rule, the water adds about 1.2 pCi/L to the air on average. That leaves roughly 4.8 pCi/L that probably comes from the soil, still above EPA's 4 pCi/L level to fix. Treating only the water would not bring this house below 4.
• 12,000 is in Maine's "10,000 or higher" band, where Maine CDC highly recommends a mitigation system. New Hampshire's guidance would also recommend water treatment together with air mitigation.
• Reasonable plan: a registered radon service provider for both an air system and whole-house aeration (Maine "typically" recommends carbon only under 10,000), then retest air and water.

House B, Vermont, drilled well. Air: 2.5 pCi/L. Water: 3,000 pCi/L.

• The water adds about 0.3 pCi/L, a small part of 2.5. 3,000 is below Vermont's 4,000 advisory level: no need to treat the water; test again in five years. The air result is in the 2 to 4 pCi/L range where EPA says to consider fixing. If anything gets fixed here, it's the soil gas.

House C, Connecticut, drilled well. Air: 1.5 pCi/L. Two water samples average 7,000 pCi/L.

• The water adds about 0.7 pCi/L to the air, nearly half of this home's 1.5.
• Connecticut says to reduce radon in water at an average of 5,000 or more, and 7,000 is within the range where Connecticut says carbon works (below 10,000). Under New Hampshire's guidance the same numbers would not call for treatment, because the air is below 4. Your state changes the answer.

When to test again

Schedules differ, and Maine's own pages differ from each other. Vermont: radon in water every five years. New Hampshire: below 4,000, retest air and water every three to five years. Maine CDC's private well page says every 3 to 5 years, its radon page every 2 years for results below 4,000, and the state lab's test guide every 5 to 10 years; the shortest interval is the cautious choice. For well testing in general (not radon specifically), Maine's private well page also lists times to test off-schedule, including "when buying a new home," after installing a treatment system, and after any work on the well.

Questions people ask