As far as state and federal officials are concerned, the drinking water in Smithwick, Texas, is perfectly safe.
Over the past two decades, the utility that provides water to much of the community has had little trouble complying with the Safe Drinking Water Act, which is intended to assure Americans that their tap water is clean. Yet, at least once a year since 2019, the Smithwick Mills water system, which serves about 200 residents in the area, has reported high levels of the synthetic chemical 1,2,3-trichloropropane, according to data provided by the Environmental Working Group, an advocacy organization that collects water testing results from states.
The chemical, a cleaning and degreasing solvent that is also a byproduct from manufacturing pesticides, is commonly referred to as T*****. It has been labeled as a likely carcinogen by the Environmental Protection Agency for more than a decade. There have been few active sources of T***** since the 1990s, but its legacy lives on because it breaks down slowly in the environment.
How it got into the Smithwick Mills water supply is a bit of a mystery. There are some farms in the area, but it’s unclear whether they have used pesticides containing the chemical, and there are no known industrial sources nearby.
The T***** levels in the Smithwick Mills system are alarming to those who study water contamination. As with many chemicals, there’s limited information on T*****’s long-term effect on humans. But research involving animals shows evidence that it increases cancer risks at lower concentrations than many other known or likely carcinogens, including arsenic. Because of this, in 2017, California state regulators set a maximum allowable level for T***** in water of 5 parts per trillion. Water quality tests from the Smithwick Mills utility have revealed an average T***** level of 410 parts per trillion over the past four years — more than 80 times what would be allowed in California.
But the utility hasn’t taken any action. It doesn’t have to. The chemical isn’t regulated in drinking water by the EPA or the Texas Commission on Environmental Quality, which means neither agency has ever set a maximum allowable level of T*****. It’s not clear why Smithwick Mills was even monitoring for the chemical, though state officials said many utilities receive results for T***** as part of routine lab tests for a variety of chemicals that get reported to state regulators.
Residents said they received no notices about the high levels, which weren’t shared in the town’s annual consumer confidence reports from 2019 to 2021, the first three years T***** was recorded in its water. T***** test results appeared in the 2022 report, which the water utility sent to residents after a ProPublica reporter reached out to the company earlier this year.
Jerri Paul, who has lived in Smithwick for three years, said she’s disappointed in the lack of communication from the water system and regulators. She has little hope that Texas officials will act on their own, because the state government has generally been reluctant to expand environmental regulation.
“I just don’t see them doing something above and beyond what the feds do,” said Paul, who is a member of the Smithwick Mill Estates property owners association board. The community is made up mostly of working-class people and retirees, many of whom don’t have the resources to buy bottled water, she noted. “We’re really dependent on the federal government doing something and saying that this is a contaminant that is not acceptable.”
A representative from Corix Utilities, which operates the Smithwick Mills water system, said in a statement that the company’s review of the tests didn’t show a danger to the community’s residents and that the system is in compliance with drinking water standards.
T***** has been found in far more drinking water than just in this small Texas town. When the EPA last conducted nationwide testing about a decade ago, the chemical was detected in the water of 6 million people (though, at the time, not in Smithwick). Four million of those people were served by systems with average concentrations above California’s standard.
T***** is one of more than a dozen unregulated contaminants that have been found in the country’s drinking water. During the past decade, regulators have identified at least one of these substances at levels that could impact human health in the tap water of 61 million people, according to a ProPublica analysis of EPA data. Nearly 16 million of these people were exposed to potentially dangerous levels of possible or likely carcinogens, including T*****. And over the past 25 years, the agency has identified more than a hundred other water contaminants, including industrial and agricultural chemicals and microorganisms, that may present risks to humans. The potential health effects of these substances include developmental delays, reproductive issues and cancer.
Experts and activists say this demonstrates fundamental shortcomings in the country’s approach to environmental threats. The Safe Drinking Water Act, designed to protect the nation’s water quality, requires an extensive, multistep process before adopting new standards. Critics say the EPA has struggled to move contaminants that have been on its radar for a decade or more through this process in a timely fashion.
The EPA’s inaction on these chemicals “just illustrates how broken the system is,” said Erik D. Olson, a lawyer who worked at the EPA during the Reagan administration and is now senior strategic director for health and food at the Natural Resources Defense Council, an environmental advocacy group. “The law really is incredibly cumbersome and difficult, and there’s just a lack of political will to regulate a lot of these things.”
An EPA spokesperson said in a statement that while the agency views T***** as a potential contaminant of concern, it hasn’t collected enough data on it. Before regulating a new contaminant, the agency must show that doing so will provide meaningful health benefits based on the law’s criteria.
“EPA must make regulatory determinations based on the best available data and peer reviewed science,” the spokesperson said in a written statement. The agency did not make officials available for interviews.
Action Is Rare
In 1974, soon after expanding regulations for surface water pollution through the Clean Water Act, Congress passed the Safe Drinking Water Act, directing the EPA to protect the nation’s tap water. But within a decade, many lawmakers felt the legislation hadn’t done enough. In 1986, Congress passed amendments to the law that directed the agency to regulate more than 80 additional contaminants, including bacteria, viruses and chemicals such as cyanide and PCBs, within five years; the EPA would have to add another 25 contaminants every three years after that.
The agency struggled to comply with the mandates and missed deadlines for setting standards. Many small water utilities and some states said that the EPA’s rulemaking process didn’t prioritize contaminants with the greatest health risks. So, in 1996, following this pushback, Congress amended the law again with the stated goal of basing the agency’s rulemaking process on “sound science.”
The amendments created a much more complex, multistep process for regulation proposals. The EPA would need to demonstrate not only that a contaminant was a danger to human health, but that it was found widely enough in drinking water to warrant regulation. The agency also had to show that the benefits of regulating the contaminant would outweigh the costs — a tricky calculation that requires the agency to weigh the known tangible price associated with treatment and cleanup versus often uncertain projections about the health impacts of newly studied substances.
“The activities of Federal agencies would not, as some have said, grind to a halt,” Republican Sens. Orrin Hatch of Utah and Jon Kyl of Arizona assured Americans in a New York Times op-ed in 1995 as the amendments were being debated.
Since then, the EPA has reviewed data on more than 35 unregulated contaminants, including sodium and the explosive RDX, through the primary process laid out in the 1996 amendments. None have yet been regulated.
In the vast majority of those cases, the agency decided there wasn’t enough evidence that the benefits of regulating a contaminant outweighed the costs. In one case — the chemical perchlorate — the agency initially decided in 2011 that it would set a maximum level, before reneging. (A federal appeals court recently ordered the agency to go through with its rulemaking process and set a standard for this chemical.)
The EPA has developed other regulations since 1996, including mandated treatment techniques and revisions to existing standards, the agency said in its statement to ProPublica. It also followed specific directives Congress made through the amendments to set limits for a handful of new contaminants using the law’s required cost-benefit analysis.
Steve Via, director of federal relations at the American Water Works Association, which represents utilities, said the agency is right to carefully consider costs before adopting new standards. Unnecessary regulations, he said, add a burden on systems that could lead to significant rate increases for customers.
“We need to protect public health, but we need to focus available resources,” he said, noting that the EPA was justified in not regulating some contaminants that weren’t widespread. “The best way to make that call is through a benefit-cost analysis.”
One family of chemicals has caused such an outcry that the streak could end soon. The EPA proposed this year to regulate a small group of perfluoroalkyl substances, or PFAS, also dubbed “forever chemicals.” The substances, which by some estimates number in the thousands and which got their nickname because they may may persist for centuries in the environment, were used in firefighting foam on military bases and nonstick materials like frying pans. They first garnered mainstream attention in the 2000s when residents in Parkersburg, West Virginia, sued DuPont, alleging the company knew that the chemicals it used at its Teflon plant there were toxic and had still exposed workers, livestock and locals to them. The company settled the lawsuit, which was portrayed in the 2019 film “Dark Waters.”
Studies have shown that prolonged exposure to PFAS in water may lead to cancer, decreased fertility, developmental delays in *****ren, immune system suppression and other adverse health effects.
The agency first gathered data on the prevalence of six PFAS chemicals from 2013 to 2016, during the same time it was testing for T*****. It found at least one PFAS chemical in the water of 17 million people, according to an analysis of EPA data.
It turns out that was a vast underestimate, in large part because the tests used at the time weren’t sensitive enough to detect PFAS at very low concentrations. Follow-up testing has uncovered additional contamination: A 2020 study based on data from all 50 states estimated that the chemicals were likely present in the water of more than 200 million people.
Amid this heightened scrutiny, the Biden administration committed to take action, leading the EPA to announce in March that it would limit six chemicals from the PFAS family. For those who have been pushing for stricter drinking water standards, the proposal has provided some hope that the agency will act on other tap water threats, though this situation was unique because of the public scrutiny around the chemicals in recent years.
Waiting for Action
Of the more than 60 other “contaminants of concern” the EPA has identified, about 20, including T*****, are possible or likely carcinogens, and nearly 30 may have reproductive and developmental impacts.
For many of those contaminants, however, there is still uncertainty about the exact human health impacts. Scientists can’t do randomized controlled experiments on humans — the gold standard used to establish cause and effect — because it is unethical to expose people to substances that might cause serious health issues. Instead, human health data typically comes from observational studies, in which researchers recruit volunteers and follow their health outcomes over time. But these are expensive, difficult to conduct and come with their own uncertainties because they are not perfectly controlled experiments.
As an alternative, researchers often turn to controlled studies conducted on rodents or other animals to project what the effect might be in humans. In the case of T*****, researchers identified a link between the chemical and cancer in mice and rats in the 1990s, but to date no large-scale studies have investigated its effect in humans.
“A lot of times people who are not trained formally as scientists or researchers hear those uncertainties up front and say, ‘Oh well, this isn’t good enough, we need to wait,’” Sydney Evans, senior science analyst at EWG, said of findings on the health effects of T*****. “One of the issues with the way that contaminants and chemicals are regulated, especially drinking water contaminants, is that it takes entirely too long. And in the meantime, so many people are being exposed, just because we can’t be 100% certain.”
There is also limited information on the contaminants’ prevalence. The EPA has collected national drinking water data on less than half of its list of contaminants, and it can only monitor for 30 of them every five years. Some advocates for increased drinking water regulation say this limit, which was part of the 1996 amendments, makes it hard for the EPA to stay on top of emerging threats.
Not every small water system is required to participate in each testing round, and even among those that do, the data collected may not be useful to regulators. For example, during the monitoring period for T*****, the lab tests the EPA directed utilities to use couldn’t detect the chemical at low levels, similar to the testing sensitivity issue the agency faced in monitoring for PFAS. In 2022, the agency demurred on taking action on T*****, in part because it had no data on how widespread the chemical was at these lower levels. The agency declined to comment on why it didn’t use more sensitive tests that were available.
As with PFAS, follow-up testing by states and local utilities have found more people exposed to T***** than was initially documented, according to the Environmental Working Group data. ProPublica’s analysis of the data shows that since the EPA stopped its monitoring, T***** has been found in water systems serving an additional 6 million people, though many states recorded few or no tests during that time period. While many of these detections were in California, which requires testing, Texas documented T***** contamination in water from Smithwick Mills and dozens of other utilities.
Alan Roberson, executive director at the Association of State Drinking Water Administrators, said the EPA should make a greater effort to provide final determinations on these contaminants of concern, including decisions to take them off its candidate list. There are three chemicals, for example, that have been on the list since 1996 without any final determination, according to the ProPublica review.
“They need a process for having a more manageable list and then doing the research to move it forward,” Roberson said. “Let’s make sure we have the stuff we need to make decisions, either up or down, on a regular basis.”
Leadership Void
In the absence of direction from the federal government, some states have acted on their own. In 2018, New Jersey joined Hawaii and California in regulating T*****. The limits vary widely, however. Hawaii’s T***** standard, which was enacted in the 1980s and revised 20 years ago, allows up to 600 parts per trillion in water.
Darrin Polhemus, deputy director of the division of drinking water with California’s State Water Resources Control Board, said the state’s laws allow it to be more aggressive in targeting health risks in drinking water. Unlike the EPA, which has to determine that the benefit of a drinking water standard outweighs the cost, California regulators are directed by state law to set a maximum level as low as possible, so long as most water systems can afford to implement the treatment.
“That is why I like our system better than the federal government’s,” he said. “It can be incredibly hard to calculate the benefit of the health outcome.”
If Smithwick Mills had been in California, the water utility would have had to drastically reduce the levels of T***** in its water to comply with the state’s standard, either by installing treatment technology to remove the chemical or changing its water supply. At minimum, residents would have been notified of the contamination levels. But since the system is in Texas, the chemical’s presence went largely unnoticed until now. State officials said they have no plans to regulate T*****.
Paul, the Smithwick resident, said what’s most unsettling is that no one seems to know how T***** entered the community’s water supply. For years, Paul drank only bottled water because she didn’t like the taste of what came from the tap. But after learning about her town’s T***** test results, she stopped giving her dog water from the tap, and now uses bottled water even to make bird feed and water her plants. She uses tap water only for cleaning and bathing.
“I don’t trust it for anything else,” she said.
Clarification, Nov. 6, 2023: The figures in the second graphic have been expanded to include the tenths decimal place so it’s more clear why the two sides are different sizes.
Maya Miller and Max Blau contributed reporting.
