Skip to main contentSkip to navigationSkip to navigation
‘The EPA is doing the bare minimum it can and that’s putting people’s health at risk,’ said Kyla Bennett, policy director at the advocacy group Public Employees for Environmental Responsibility.
‘The EPA is doing the bare minimum it can and that’s putting people’s health at risk,’ said Kyla Bennett, policy director at the advocacy group Public Employees for Environmental Responsibility. Photograph: Massimo Percossi/EPA
‘The EPA is doing the bare minimum it can and that’s putting people’s health at risk,’ said Kyla Bennett, policy director at the advocacy group Public Employees for Environmental Responsibility. Photograph: Massimo Percossi/EPA

Revealed: US water likely contains more ‘forever chemicals’ than EPA tests show

This article is more than 1 year old

Guardian analysis of water samples taken in nine US locations shows test agency uses is likely missing significant levels of PFAS pollutants

A Guardian analysis of water samples from around the United States shows that the type of water testing relied on by the US Environmental Protection Agency (EPA) is so limited in scope that it is probably missing significant levels of PFAS pollutants.

The undercount leaves regulators with an incomplete picture of the extent of PFAS contamination and reveals how millions of people may be facing an unknown health risk in their drinking water.

The analysis checked water samples from PFAS hot spots around the country with two types of tests: an EPA-developed method that detects 30 types of the approximately 9,000 PFAS compounds, and another that checks for a marker of all PFAS.

The Guardian found that seven of the nine samples collected showed higher levels of PFAS in water using the test that identifies markers for PFAS, than levels found when the water was tested using the EPA method – and at concentrations as much as 24 times greater.

“The EPA is doing the bare minimum it can and that’s putting people’s health at risk,” said Kyla Bennett, policy director at the advocacy group Public Employees for Environmental Responsibility

Ties to cancer

PFAS are a class of chemicals used since the 1950s to make thousands of products repel water, stains and heat. They are often called “forever chemicals” because they don’t fully break down, accumulating in the environment, humans and animals. Some are toxic at very low levels and have been linked to cancer, birth defects, kidney disease, liver problems, decreased immunity and other serious health issues.

The Biden administration in June announced new actions aimed at protecting drinking water from PFAS contamination, saying the chemicals “pose a serious threat across rural, suburban and urban areas”. The administration has allocated $10bn to specifically address PFAS and other contaminants in drinking water.

But critics say when it comes to identifying PFAS-contaminated water, the limitations of the test used by state and federal regulators, which is called the EPA 537 method, virtually guarantees regulators will never have a full picture of contamination levels as industry churns out new compounds much faster than researchers can develop the science to measure them.

That creates even more incentive for industry to shift away from older compounds: if chemical companies produce newer PFAS, regulators won’t be able to find the pollution.

“Industry has had a 70-year head start and we’re never going to catch up,” said Graham Peaslee, a University of Notre Dame researcher.

chart showing higher PFAS levels using the TOF test vs EPA 537

Many researchers consider a type of test known as “total organic fluorine” (TOF), which detects a PFAS marker called organic fluorine, to be the most accurate way to test water samples.

The European Union is proposing switching to a TOF test, and states such as Maine, which are planning to regulate PFAS as a class instead of regulating individual compounds, will need more robust testing to enforce their laws.

“The TOF isn’t the end-all, but it’s telling us there’s more fluorine out there, and we need to look for it,” Peaslee said.

Clean water advocacy groups last year urged the EPA to use more comprehensive tests that they said would “give us a better understanding of the totality of PFAS contamination”, but the agency told the Guardian it currently has no such plans.

In a statement to the Guardian, the EPA said it “continues to conduct research and monitor advances in analytical methodologies … that may improve our ability to measure more PFAS”.

For researchers worried about PFAS contamination, that is not good enough.

“We’re looking for and studying less than 1% of PFAS so what the heck is that other 99%?” Peaslee asked.I’ve never seen a good PFAS, so they’re all going to have some toxicity.”

Guardian sampling

The samples analyzed for the Guardian were collected from municipal systems and private wells, including both filtered and unfiltered water. An accredited lab conducted the EPA 537 test, while Peaslee checked the samples using a TOF method he developed.

In unfiltered water from Portsmouth, New Hampshire, the levels were 10 parts per trillion (ppt) in the EPA 537 test and 164ppt in the TOF test.

Water samples collected in Titusville, Florida, also showed a large disparity – the EPA 537 test found PFAS at 16ppt, while the TOF test found PFAS levels at 176ppt. In Bethesda, Maryland, the results were 18ppt from the government-favored test and 185ppt from the TOF.

Similar results were seen in sampling from other communities, including in Wisconsin, North Carolina, Arizona and Massachusetts.

Though the EPA doesn’t have limits in place for mixes of PFAS compounds, many public health advocates say no level over 1ppt is safe.

One of the samples in the Guardian analysis – water from Oscoda, Michigan – showed 13.7ppt in the EPA 537 test and 0 (non-detect) for the TOF test. Testing of a sample from Gustavus, Alaska, found 127ppt in the EPA 537 test, and a lower amount – 102 ppt – in the TOF test, considered within the margin of error.

The EPA and industry have long argued that many newer PFAS that can’t be detected are safe. However, most new compounds have not been independently reviewed, and the types of PFAS that have been studied have been found to be toxic and persistent in the environment, said Bennett.

“There are so many PFAS that we don’t know anything about, and if we don’t know anything about them, how do we know they aren’t hurting us?” Bennett asked. “Why are we messing around?”

The test results for Portsmouth, where water tested by the TOF method revealed levels around 16 times higher than the EPA 537 method found, is probably due to some combination of issues, Peaslee said.

Though firefighting foam used at Pease air force base in Portsmouth and elsewhere is largely made with PFAS compounds the EPA test can detect, the chemicals, once in the environment, break down into different PFAS compounds that can’t be detected.

It’s also possible that other sources are polluting the region’s water with newer PFAS that can’t be read by the EPA test.

The results are “surprising”, said Andrea Amico, a public health advocate who in 2014 first sounded the alarm about Portsmouth’s PFAS contamination, and who collected the water sample from Portsmouth used in the Guardian analysis.

“That’s left me with more questions about what’s making up that total and makes me want more testing in my community,” she added.

Health problems

In the region around Cape Canaveral, Florida, which includes Titusville, some suspect PFAS contamination stemming from two military bases and Nasa facilities is behind their health problems.

Since 2019, Titusville utility officials have either reported no PFAS in the city’s drinking water or have said detections were at levels considered by regulators to be safe. But the TOF analysis for the Guardian detected 176ppt in the water there.

Among the water samples collected for the Guardian, some came from the home of a Titusville resident who suffers thyroid problems, a condition linked to PFAS exposure. The resident, who declined to be named, can’t afford a water filtration system, a situation that underscores the fact that many low-income people can be at more risk than people with higher incomes.

“They used [the EPA 537] results as cover,” said Stel Bailey, who has suffered from PFAS-linked ailments such as Hodgkin’s lymphoma and works with the clean water advocacy group Fight for Zero. “We need better testing technology so we know where to focus.”

Contamination stemming from airport and military facilities has also plagued Tucson, Arizona, for years, and new measures put in place by state and local officials are supposed to have largely eradicated the problem.

But water from the home of one south side Tucson resident sampled for the Guardian showed 2ppt in the EPA 537 test and a level 24 times higher in the TOF test. Mary Ann Granillo, the resident of the home where the sampled water came from, has lupus, a PFAS-linked disease that already killed two of her family members.

City officials disputed the veracity of the sampling, stating that south side Tucson residents have not been served water from contaminated areas near airport and military facilities since 1994.

Granillo said she can’t afford a water filtration system, and bottled water is an expensive addition to her monthly bills. The family washes dishes, cleans clothes and showers in contaminated water. She fears nothing is likely to change.

“It really worries me a lot,” she said.

This project was funded by the Society of Environmental Journalists and Fund for Investigative Journalism

This article was amended on 6 July 2022 to reflect the fact that water from Portsmouth, New Hampshire, included in the Guardian analysis was unfiltered and had not been treated at the city’s water treatment facility; references to the Portsmouth water treatment facility have been deleted. It was further amended on 7 July 2022 to add specifics about the sampling in Tucson and include Tucson officials’ disputation of the sampling’s veracity. And the graphic was updated on 12 July 2022 because the water sample from North Carolina came from Navassa, not Leland as an earlier version said.

Most viewed

Most viewed