A damaged water line can pull fire toxins straight into tap water
One utility found fire-related contamination in more than half the service connections it tested near burned lots. Boiling can make that worse by driving benzene into the air.
Table of contents
- How does a fire get into water that never burned?
- Can you boil it to make it safe?
- Does your house have to burn for this to happen?
- How bad did it get after past fires?
- Who is supposed to test the water inside your house?
- Why can a clean test come back wrong?
- What happened in Lahaina?
- How do you get your own tap tested?
Summary of this article
- Wildfire gets into tap water two ways: plastic pipes melt and release chemicals, and lines that lose pressure suck in smoke.
- Your house does not have to burn. One utility found fire-related contamination in more than half of service connections at burned lots nearby.
- Boiling can make it worse. Heat drives benzene into air you breathe, and utilities in California and Oregon told people to boil water that was chemically contaminated.
- Santa Rosa measured benzene at 40,000 micrograms per liter after the 2017 Tubbs Fire. The federal limit is 5.
- Nobody is required to test inside your house. Utilities test their own system, and everything past your meter is yours.
- A test right after flushing can read clean on a pipe that is not. Water has to stand still first.
- In Lahaina, 586 household samples turned up 14 chemicals out of 78 screened, which is what a careful check often looks like.
How does a fire get into water that never burned?
Two ways, and neither one needs a flame near your kitchen. Plastic pipe, meters and fittings break down in heat and give off chemicals into water sitting inside them. And a line that loses pressure stops pushing water out and starts pulling in, so smoke and hot gas get drawn through broken mains and open hydrants.
Picture pulling into your driveway. Half your block is gone. Your house is standing, power is back on, and you run a glass of cold water that looks entirely normal, because it does.
That glass came through pipe you share with every lot on your street, burned ones included. Melted plastic sitting upstream of your tap doesn’t know which houses made it. Nor does water.
California’s Division of Drinking Water says the hard part out loud, that “It is unknown without testing which pipes may be affected.” One street, one fire, two houses, and water can differ at each one.
It doesn’t clear itself either. Benzene soaks into walls of plastic pipe and comes back out slowly, so a pipe can hand something over weeks after a fire is out. We’ll come back to that.
Can you boil it to make it safe?
Boiling can make it worse. EPA warns that until chemicals are ruled out, boiling water or showering in it “could create an inhalation risk,” because heat drives them out of water and into air you stand in.
Your phone buzzes with a boil-water notice from your utility. That notice is routine after pressure drops and it’s aimed at germs. Boiling kills those. It does nothing to benzene except move it from your kettle into your kitchen.
This has already gone wrong, and not in a small way. Whelton and colleagues, writing up Colorado’s Marshall Fire, found utilities and state agencies in California and Oregon telling people to boil water that was full of chemicals. Nobody was careless. A boil notice is the one message every utility has sitting ready to send.
The warning that matches this is a “Do Not Drink – Do Not Boil” advisory, and CDC does not recognize one, in current guidance or in its own toolbox. So the advice that fits is missing from that kit, and the wrong one is a click away.
If we could get one line from this page into a neighbor’s head, it would be that one. Fire through your area, pressure gone, and we’d stay on bottled water until somebody samples.
Does your house have to burn for this to happen?
No. One utility found fire-related contamination in over 50 percent of service connections at burned lots, and houses that didn’t burn drink from that same network.
Colorado, December 2021. The Marshall Fire destroyed 1,084 buildings and damaged 149 more across Louisville, Superior and Boulder County. Six public water systems took damage. Every one of them lost power.
Pressure dropped in Louisville, in East Boulder County and in Superior. Two of those three came back with chemicals in their water. Superior lost buildings and pressure both, and its network came back clean, which is worth knowing before anybody tells you this always happens.
Most homes in those neighborhoods were still standing. Every one of them was fed by a system that had been through all of that, the way yours would be.
We’d read it this way. Whether your house burned tells you about your house. It says very little about water arriving at it, which comes from somewhere else, through pipe you don’t own and can’t see.
How bad did it get after past fires?
Bad enough that people don’t believe it first time. Santa Rosa’s water carried benzene at 40,000 micrograms per liter after the 2017 Tubbs Fire. The federal limit for benzene in drinking water is 5.
Forty thousand, against five. Read that twice.
Those results stand in a table Whelton and colleagues built for AWWA Water Science, covering systems hit by fires between 2017 and 2020. Paradise Irrigation District came in above 2,217 after the 2018 Camp Fire. Researchers logged 55 different chemicals across those systems, and eleven had gone past drinking-water limits after earlier fires.
The National Academies went back over Tubbs and Camp and found benzene above 5 parts per billion in mains, hydrants and service connections. They put it down to burning household materials, pulled into pipe wherever the inside fell below the outside, and it hung around for months until crews cleaned it out.
We’re not printing this to frighten anybody, and it isn’t what your street should expect. Most systems come back fine after most fires. California’s State Water Board cleared the last of nine systems hit by January 2025 Los Angeles firestorms that May, advisories and all. Our point is narrower, and we’d rather you had both halves. Nobody modeled this on a computer. It happened, in towns like yours, more than once.
| Who owns it | Who samples it | |
|---|---|---|
| Source water and treatment plant | the utility | the utility |
| Mains under the street | the utility | the utility |
| Service line to your meter | shared | the utility, to the meter |
| Everything past the meter | you | nobody |
Who is supposed to test the water inside your house?
Nobody, on most streets. EPA states that plumbing from your meter inward belongs to whoever owns the house, that utilities have no standing requirement to sample inside buildings, and that no agreed method for doing it exists.
Three gaps stacked on each other, and between them they decide whether you ever find out.
Your utility tests what it owns. Mains, hydrants, the network under your street. When that comes back clean it says so, and it’s telling the truth about the part it tested.
Everything past your meter is yours, no rule says anybody has to sample it, and nobody has settled on how that would even be done if somebody volunteered.
So you hear that a water system has been cleared and take that to cover your kitchen. Fair enough. It covers your street. That gap is the whole point of this page.
California wrote a law about it, and it’s worth knowing how far that law reaches. AB 541 makes a public water system sample for benzene after a wildfire of 300 acres or more that damaged or destroyed a building hooked up to it. Lawmakers put Santa Rosa and Paradise into the bill’s own findings. Then look at what gets sampled. Source water, treatment plants, pipes between them, mains under a street. All of it’s the utility’s own gear. Even that law stops at your meter.
Why can a clean test come back wrong?
Because of whatever happened in that pipe right before somebody filled a bottle. Benzene comes out of plastic slowly, so water has to stand still long enough to pick any up. California’s Division of Drinking Water tells residents to leave plumbing untouched for at least 8 hours before sampling, and after past fires waits of 8, 24, 48 and 72 hours have all been used, with longer ones pulling more out of a pipe.
Boulder County again. A customer at East Boulder County Water District rang to say their water smelled of paint thinner. The district flushed that line, took a sample, and it came back with nothing.
Weeks later they sampled again, this time letting water stand first. Benzene came up at 0.8, 1.0 and 5.8 micrograms per liter in first-draw samples, and other chemicals came with it. Flushed samples stayed clean.
Nothing went wrong at that lab. Flushing pushes fresh water through a pipe, so a sample taken straight after measures the flush. Let water stand, and whatever soaked into plastic starts coming back out.
So your result is only as good as whatever came before it. Somebody runs your taps that morning, or flushes your line to be helpful, and an afternoon sample says clean on a pipe that isn’t.
We would ask whoever tests your water how long it stood still first. If the answer is that they flushed and then sampled, that result describes the flush, and we’d want to know that before paying for it.
What happened in Lahaina?
Mostly good news, and it came from residents filling bottles at their own kitchen taps. A community-run program collected 586 samples from more than 450 households in Lahaina and Kula after the August 2023 Maui wildfires, screened for 78 chemicals, and found 14.
Those fires destroyed or damaged over 2,200 buildings. Sampling ran more than a year, in people’s homes, by people who lived there, while Maui County’s water department checked its sources and public mains and found nothing fire-related. This program went after the half nobody checks, which is the plumbing inside a house, past your meter.
Biggest single result was tetrahydrofuran at 216 micrograms per liter. Chloroform topped out at 86.3. One thing crossed a federal limit, total trihalomethanes at 95.2 against 80. Those come from chlorine in water rather than from a fire. McLatchy and colleagues put their conclusion carefully, saying residents in their study had “very limited to no drinking water contamination.”
We wanted this in because it cuts the other way. 14 out of 78 isn’t nothing, and those researchers didn’t call it nothing. It is also nowhere near Santa Rosa. We think both directions belong on one page, because a careful check can end right here. A few things present, nothing awful, and now you know.
So Santa Rosa is the high end of what anybody has measured. Most streets look more like Lahaina, and a test is what tells you which one you’re on.
How do you get your own tap tested?
Leave your taps alone overnight, take a cold-water sample at your kitchen faucet in the morning, and send it to a state-certified lab. Water has to stand still first. Somebody running a shower at six undoes the test you paid for.
Flushing isn’t the enemy of testing. Taking a sample straight after a flush is. California’s State Water Board tells residents to flush cold taps about 5 minutes each, run hot taps until water turns cold, swap out refrigerator and under-sink filters, and clean out ice maker lines. Then leave everything alone overnight before that morning sample.
Ask for a wide chemical scan, not benzene alone. EPA says outright that benzene isn’t a perfect indicator, because samples free of benzene have carried other chemicals.
Then replace whatever was filtering bad water. Fridge cartridges. Under-sink units. Ice makers. A filter that caught something is still full of it, and we wouldn’t leave one in a house somebody has only now moved back into.
One last thing, plainly. A fire that never reached your house can still reach pipe feeding it, and no rule anywhere requires anybody to check that pipe on your side of a meter. Whether your kitchen ever gets tested comes down to which state you live in and whether you asked. That isn’t how most people assume drinking water works.
Sources
- U.S. Environmental Protection Agency, Office of Water. Addressing Contamination of Drinking Water Distribution Systems from Volatile Organic Compounds (VOCs) After Wildfires (2021) EPA 817-F-21-011
- Whelton AJ, Seidel C, Wham BP, Fischer EC, Isaacson K, Jankowski C, MacArthur N, McKenna E, Ley C. The Marshall Fire: Scientific and policy needs for water system disaster response, AWWA Water Science (2023) AWWA Wat Sci 5(1):e1318
- California State Water Resources Control Board, Division of Drinking Water. Information to Water Customers Regarding Water Quality in Buildings Located in Areas Damaged by Wildfire (2019)
- California Legislature. Assembly Bill 541, California Safe Drinking Water Act: wildfire aftermath: benzene testing (2023) Chapter 530, Health and Safety Code section 116596
- California State Water Resources Control Board. L.A. firestorms recovery milestone: Safe drinking water restored for all impacted public water systems (2025)
- U.S. Environmental Protection Agency. 40 CFR 141.61, Maximum contaminant levels for organic contaminants (2023)
- U.S. Environmental Protection Agency. 40 CFR 141.50, Maximum contaminant level goals for organic contaminants (2023)
- California Office of Environmental Health Hazard Assessment. Public Health Goal for Benzene in Drinking Water (2001)
- California State Water Resources Control Board. MCLs, DLRs, and PHGs for Regulated Drinking Water Contaminants (2025)
- McLatchy LA, Whelton AJ, Rong KC, Cole KDP, I'i J, Shuler CK. Community-Driven Water Quality Assessment Following the 2023 Maui Wildfires, ACS ES&T Water (2026) ACS EST Water 6(2):793-804
- Rowe BL, Toccalino PL, Moran MJ, Zogorski JS, Price CV. Occurrence and Potential Human-Health Relevance of Volatile Organic Compounds in Drinking Water from Domestic Wells in the United States, Environmental Health Perspectives (2007) EHP 115:1539
- National Academies of Sciences, Engineering, and Medicine. The Chemistry of Fires at the Wildland-Urban Interface, chapter 5, Water and Soil Contamination (2022)
- Building Biology Evaluation Guidelines. SBM-2024 Standard Points B3-B6, Solvents and other Volatile Organic Compounds (2024)
Questions people ask
Can a wildfire contaminate the water in a house that did not burn?
Yes. EPA's fact sheet EPA 817-F-21-011 states that VOC contamination can be present without visible damage, because pipes and meters in a fire area may still hold pressure and flow. It also states that the only way to determine whether the water has elevated VOC levels is to collect samples for analysis.
How does benzene get into drinking water after a fire?
Two proposed mechanisms, per EPA. Plastics such as PVC and high density polyethylene in pipes, meters and gaskets degrade when heated and release chemicals. And when water pressure drops, smoke, hot gases and chemicals from burning vegetation and structures are drawn into the lines.
Is it safe to boil the water after a wildfire?
No, not until chemical contamination has been ruled out. EPA writes that boiling the water or hot water use such as showering could create an inhalation risk while VOC contamination is possible, and recommends a Do Not Drink, a Do Not Drink - Do Not Boil, or a Do Not Use advisory instead of the usual boil water notice.
How long does water have to sit before a post-fire sample is valid?
At least eight hours. California's Division of Drinking Water tells residents to flush, then leave the plumbing untouched for a minimum of eight hours before sampling. EPA notes that stagnation periods of 8, 24, 48 and 72 hours have been used after past fires, with longer periods giving more chemical time to leach back into the water.
What is the limit for benzene in drinking water?
The federal maximum contaminant level is 5 micrograms per liter, written as 0.005 mg/L at 40 CFR 141.61. That is the enforceable figure. The federal maximum contaminant level goal is zero, California's state limit is 1 microgram per liter, and California's public health goal is 0.15.
Does anyone test the plumbing inside my house after a fire?
No agency is required to. EPA states that property owners are responsible for the service connection from the meter and all indoor plumbing, that utilities are not as a general matter required to sample inside structures, and that no standards or accepted protocols for sampling homes exist.
Does a wildfire always contaminate the water?
No. After the 2023 Maui fires, a community-run study of 586 samples from more than 450 households found that only total trihalomethanes exceeded a federal limit, and those come from chlorination. The authors concluded that residents in their study experienced very limited to no drinking water contamination.
Have questions about the environment inside your home?
Air quality, mold, EMF, light and water all shape how a house feels and how the people in it feel. Whatever your concern is, the first step is finding out what is there.
Our certified team of environmental professionals and advocates can help.


