A water expert discusses a dramatic shift in federal oversight
It didn’t grab headlines, but the U.S. Environmental Protection Agency’s (EPA) decision last month to back away from regulating a rocket fuel ingredient in drinking water points to a dramatic shift in federal oversight.
The decision was followed by a proposal to slow the process for reviewing chemicals under the Toxic Substances Control Act (TSCA) and delayed action on hazardous perfluorochemicals, PFAS and PFOA, that have been found in various water systems.
Below, Richard Luthy, the Silas H. Palmer Professor of Civil and Environmental Engineering at Stanford, discusses how the EPA and individual states regulate water, risks for certain communities and related issues.
The EPA has signaled its intention to slow down its review process for chemicals under the Toxic Substances Control Act (TSCA). To what extent have chemicals under this act been of concern for water supplies?
TSCA requires reporting of chemicals in commerce to ensure that they are being used safely. The drinking water industry, which has the burden of removing chemicals that may adversely impact drinking water sources, wants the EPA to use its authority under TSCA to prohibit or restrict the use of such chemicals. A better approach is preventing contaminants from entering drinking water sources in the first place.
How could a slowdown in the Toxic Substances Control Act review process affect water systems? About how long should the review process take?
TSCA was revised in 2016 and requires the government to review many chemicals at any time until all chemicals have a safety determination. It’s a daunting task to make headway in assessing tens of thousands of chemicals in use, accumulating all the data, amassing the people to do the work and preparing for challenges in determining unreasonable risk. The revised TSCA sets a 3-and-a-half-year period for establishing a list of 20 high-profile chemicals for risk evaluation. Even absent challenges and responses, you can see how long it would take to work through a backlog of chemicals. Presently, environmental groups are suing the EPA over lack of transparency and public input.
Experience shows that it might take 20-30 years or more after a chemical has entered the commercial market to strictly regulate or prohibit its use, as in the case of lead, PCBs and DDT. It’s hard to speculate to what extent the EPA’s new, slower process might affect this timeline because administrations and priorities change.
Why is EPA missing deadlines for chemical reviews? Specifically, can you discuss the case of PFOA and PFAS?
Although there are no enforceable federal drinking water standards, the EPA has established health advisories for perfluorochemicals. These chemicals are being phased out from consumer products such as nonstick coatings and stain-resistant treatments. Their use in industry or aqueous fire-fighting foams resulted in groundwater contamination, notably at fire-fighting training facilities and military bases. States like California have moved ahead of the EPA to regulate perfluorochemicals.
Unfortunately, the current trend seems to be one of the EPA acting very quickly to repeal regulations, while taking time to protect drinking water. An example is methyl tertiary butyl ether (MTBE), a once-used gasoline additive that was found to contaminate groundwater from gas stations. MTBE still has not been regulated at the national level. Meanwhile, the administration is rolling back its definition of what water bodies deserve protection under the Clean Water Act.
What are the chances of being exposed to perchlorate in water? Is it more likely to be in certain regions or communities? How do we know?
Perchlorate is not commonly found in drinking water supplies. Perchlorate is used in the manufacture of pyrotechnics, highway flares and solid rocket fuel. It is found predominately in soil and groundwater where the manufacture of oxidizers for missiles and rocket propellants took place historically.
Surveys in the U.S. show perchlorate in water in Southern California — mostly in Los Angeles, San Bernardino and Riverside counties — west Texas and along the East Coast from New Jersey to Massachusetts, potentially impacting water to about 10 million people. That said, exposure is decreasing with time due to cleanup and provision of treated or alternative drinking water supplies. We know this because perchlorate has been studied for regulation in drinking water for about 20 years, so there’s a lot of monitoring data available.
How does the EPA decide which waterborne contaminants to regulate and how? Why did perchlorate fail that test?
The EPA considers three criteria: Does the contaminant cause adverse health effects if present in drinking water? Is there a substantial likelihood that a contaminant will occur with a frequency and at levels of public health concern? Does the regulation of the contaminant present a meaningful opportunity for health risk reduction?
The main argument against regulating perchlorate is related to the question of frequency and levels. Perchlorate is found only in specific locations, as I noted earlier. Also, there is uncertainty about the level of health concern, whether two or six parts per billion — as regulated in Massachusetts and California — or 56 parts per billion, as was proposed by the EPA.
Are state legislators and municipal utilities doing enough — without EPA oversight — to ensure perchlorate doesn’t reach customers?
States are ahead of the EPA on regulation of perchlorate in drinking water. California is considering whether the goal should be 1 part per billion based on reduced uptake of iodine into the thyroid gland by infants. Unfortunately, there’s quite a bit of variance in risk reduction calculations and whether vulnerable populations are considered.
Why did water utility groups support the EPA move?
Utilities argued that, because perchlorate wasn’t frequently found across the country, widespread testing for it is a waste of money. Regardless, there is always the option to monitor for several years and then seek a waiver for further testing. This would seem the more prudent course of action.