Problems in the pipes: Water woes stir concerns of wider contamination
Feb. 15, 1998
Michael Mansur, environmental writer
© 1998 The Kansas City Star
TROY, Kan. -- Six years ago, Kansas regulators tested Julie Speer's water and found stunning levels of a cancer-causing chemical.
Over the years, they tested again and again. Each time, they found the contaminant at levels above federal limits. Even top water regulators in Washington, D.C. knew about the Speer family's contaminated water.
But no one told Speer of the high chemical levels in her family's water until last October.
"If they knew about this back in '93 and had told me, I would have done something back then -- instead of waiting until now," said Speer, who has two children: Tristan, 1, and Gabrielle, 2.
Regulators did not act to reduce their exposure to the cancer-causing chemical until The Kansas City Star began to investigate the contamination. This week, the Speer family and about two dozen other northeast Kansas families will start receiving bottled water.
Tests of their water found vinyl chloride, known to cause cancer in humans, at levels up to seven times what the U.S. Environmental Protection Agency has set as safe.
Doniphan County's water contamination is the nation's first confirmed case of polyvinyl chloride, or PVC, pipe tainting drinking water at levels above the federal limit, EPA officials say.
But Doniphan County's problem raises concerns about the potential for a much larger chemical contamination problem in hundreds of rural water districts that distribute potable water across the nation's rural midsection, from the Dakotas to Texas, The Star's investigation has found.
Many of those systems, like Doniphan County Rural Water District No. 5, use a form of PVC pipe manufactured before 1975 that the pipe industry now knows can leach, or release, vinyl chloride into drinking water under some conditions.
Urban systems like Kansas City's generally do not have vinyl chloride contamination problems. And while PVC pipe was used in many suburban developments, contamination is not believed to be an issue there, pipe producers said. Water sits in the pipes much longer in rural systems that are far-flung but have comparatively low consumption.
But hardly any of those rural water districts have ever been checked for vinyl chloride contamination, federal and state regulators acknowledge.
Many of those systems -- which number more than 1,800 in eight states -- also contain newer sections of PVC pipe that shouldn't leach significant levels of the chemical. The industry changed the way it made the pipe after it learned, in the mid-1970s, of vinyl chloride's cancer potency.
"My hope is we find this is an isolated problem," said Bob Morby, who heads the drinking water office in the EPA's regional office in Kansas City, Kan. "That would be best for everybody.
"If not, we've got to find out how widespread it is and how we can deal with it."
Morby said it's clear now that four years ago the Doniphan County contamination case "fell through the cracks" among the EPA's headquarters and regional offices and the Kansas Department of Health and Environment.
"We can all look back and ask, `Did headquarters drop it? Did the state drop it? Did the region drop it?'" Morby said.
"I mean, there's probably enough (blame) to go around for us all."
Attention should be focused now, he said, on how to reduce the exposure of Doniphan County residents or anyone else who might be exposed to vinyl chloride in their drinking water.
Scientists who specialize in the health threats of vinyl chloride and similar contaminants suspect Doniphan County residents exposed to the carcinogen may face only a minor increased risk of cancer.
But almost all agree that it's best to avoid exposing people to any chemical that may pose an increased cancer risk, even if it's minimal. Federal law sets as its goal the elimination of all carcinogens from drinking water.
Vinyl chloride producers say Doniphan County residents exposed to vinyl chloride face an insignificant risk.
Pipe industry experts have determined that some Doniphan County residents probably have been exposed to the chemical since the pipe's installation nearly 30 years ago.
Doniphan County Rural Water District No. 5 sent notices to customers in 1992, 1993 and 1994 that vinyl chloride had been detected in the water system. But individuals with high levels of the cancer-causing chemical in their water, such as Speer, didn't receive specific warnings and test results until last October.
The first notice stated that "the Kansas Department of Health and Environment does not believe these levels to be a serious health risk in a short time exposure," but could act as a carcinogen under extended exposure. Follow-up notices did not address the seriousness of the risk.
As soon as she received the notice with test results of her own home's water, Speer began buying bottled water.
"I've seen a lot of people die from cancer," Speer said. "And whenever there's a chance to try and prevent it, you want that opportunity."
No tap testing
Regulators first learned by accident that vinyl chloride contaminated Doniphan County Rural Water District No. 5.
In 1989, the nearby farm town of Bendena, Kan., needed to dilute its water to reduce the levels of nitrates in it. So Bendena bought some of District No. 5's water.
Kansas regulators tested Bendena's blended water for nitrates. But strangely, they detected vinyl chloride. They found little of the contaminant when it entered District No. 5's system. But they found high levels within the pipe system and out of residents' taps.
The highest levels were found in the system's "dead-end" sections, the farthest tentacles of the sprawled distribution system of about 100 miles of PVC pipe. It was in those "dead-end" sections that water sat the longest. The levels, Kansas regulators learned, also grew higher in warmer months.
If not for Bendena, the vinyl chloride might not have been detected because the federal Safe Drinking Water Act requires communities to test their water for purity only at the point it enters the distribution system, not from homeowners' taps.
"We were surprised at the levels (of vinyl chloride) we were seeing in Doniphan County, that they were exceeding the standard," said Steve Clark, who in the early 1990s oversaw drinking water standards for the EPA in Washington.
"There had been concerns about this over the years, but I had never seen anything like this."
Confirming contamination
When Dave Eckstein heard that PVC pipe in Doniphan County was suspected of leaching vinyl chloride, he dismissed it as another false report.
Technical adviser of the Dallas-based Uni-Bell PVC Pipe Association, which represents PVC pipe producers, Eckstein thought it was "an unbelievable suggestion."
"We had never encountered this in the field before," he said.
PVC pipe is made with "polymerized" vinyl chloride, which means the vinyl chloride molecules are linked like a chain and normally won't leach into water.
But before about 1975, pipe manufacturers used PVC resins that also contained unlinked molecules of vinyl chloride, known in the industry as residual vinyl chloride monomer. The flow of water through the pipe could pick up the unlinked vinyl chloride.
By August 1993, pipe producers had confirmed that the pipe was leaching vinyl chloride into drinking water in Doniphan County.
In November 1993, Ralph Flournoy, then chief of the drinking water branch of the EPA's regional office, expressed concern that the problem might extend well beyond Doniphan County.
The same type of PVC pipe, he wrote in a memo to headquarters, had been used in hundreds of other rural water districts across the nation's rural center. Flournoy suggested replacing the problem pipe or installing equipment at each house to remove the vinyl chloride from the water.
Flournoy's staff had already done some rough calculations of the risk posed to District No. 5 customers. The draft assessment concluded that "this situation does not appear to be dismissed as negligibly risky," and it recommended that residents receive bottled water.
Today EPA officials say they never acted on that recommendation because it wasn't final and the person who developed it left the agency soon after its development.
Back at EPA headquarters, the pipe industry produced for Washington's EPA staff a report that confirmed beyond doubt that sections of the Doniphan County pipe were leaching vinyl chloride into the drinking water above the EPA limits. Now the question was what to do about it.
Things fall apart
While pipe producers worked on their final study of the contamination, EPA efforts to solve Doniphan County's problem -- and to unravel the widespread nature of the contamination -- were about to crumble.
The key blow came from the EPA's Office of Ground Water and Drinking Water, which ruled in March 1994 that the vinyl chloride in Doniphan County's water was not a violation of federal drinking water law. Strictly read, the law calls for contamination to be measured at the point where the water enters the utility's distribution system. High levels detected within the system, or in water flowing from individual faucets, could not be the basis for a violation.
"If you put your blinders on and read the regulation, it does say compliance is determined at the point of entry," said Dave Waldo, who oversees public water supplies for the Kansas Department of Health and Environment.
"But those regulations didn't anticipate something like this happening."
Nonetheless, the ruling meant the EPA couldn't force the water district or the pipe industry to remedy the problem.
"In light of (the ruling), we, as an agency, did not pursue the violation any further," explained Morby, who at that time supervised another branch in the EPA's regional office. The ruling also had an effect on pipe producers.
"There's really not a violation of anything yet," Eckstein now points out. "(That) really clouds things for industry."
The decision might also have delayed dealing with the Doniphan County problem, according to state public records. When Kansas regulators pressed Eckstein on whether his group was willing to help Doniphan County replace its problem pipe, Eckstein said he'd be on-site in a week with a truckload of pipe to "make this headache go away," if he could confirm it wasn't a widespread problem, according to a memo from Waldo to his boss.
"However, without sampling of other systems, the industry is reluctant to provide financial commitment necessary to replace the pipe without knowing how many other problems exist," Waldo wrote.
Eckstein does not deny his remarks to Waldo. But he said the comment should not be considered an indication that the industry has hesitated to do the proper thing regarding Doniphan County's drinking water contamination.
Any industry representative obviously would want to know the implications of any action he took on behalf of that industry, Eckstein said. "It's about scope in a sense," he said. "What do you want to do? How far beyond where you should be do you want to step?
"I take exception to any suggestion we've hesitated. We've met every deadline and we've proposed most of them."
But soon after June 1994, the Doniphan County case, and any plans to study a widespread problem, went into the regulators' files. They took nearly four years to resurface.
A new start
Morby had only been on the job about 18 months when he first heard about the contamination in Doniphan County late last year. He quickly made it a priority.
He reassembled state regulators and industry representatives in Kansas City. He held a community meeting in Doniphan County. New testing of water in the pipe system began.
Morby informed his counterparts in other EPA regions. He contacted other Midwestern states. Drinking water regulators in Iowa, Nebraska, Kansas and Missouri have now committed to sample this spring or summer for vinyl chloride in rural water districts. Pipe and vinyl chloride producers will help pay for analysis of those samples.
EPA officials acknowledge it's possible that other rural water districts are contaminated by vinyl chloride. But they have yet to find money to support a wider testing program.
EPA knows of only one other rural water district that has shown high levels of vinyl chloride. In 1992, Kansas regulators checked 18 rural water districts they thought to be similar to Doniphan County. One showed vinyl chloride above the federal limit of 2 parts per billion; when rechecked, though, the level had fallen below the limit.
Pipe producers don't think there is widespread vinyl chloride contamination in the nation's rural water districts. But the industry expects that a few other rural water districts may be found to have vinyl chloride levels above the EPA's limit.
"That's why we're encouraging that we move forward and let's get the testing done," Eckstein said.
If many districts had contamination, Eckstein suspects his association would have picked up reports.
The industry's study of Doniphan County shows that some adverse conditions must align in order for the contamination to occur. Those conditions include pre-1975 pipe, temperatures above 50 degrees and water sitting in the pipe for long periods of time.
"If these adverse conditions line themselves up, you can get this, yes," Eckstein said. "But my belief is that these circumstances are just not going to align themselves that often. That's why I don't think there's a, quote, widespread problem."
In Doniphan County, many steps are being taken to reduce the exposure of individuals to the cancer-causing chemical.
Last October, state regulators sent individuals for the first time the analysis of the levels in their water.
"Looking back," Waldo said, "I wish we had done that (earlier)."
This week, bottled water will be delivered to up to 28 customers of District No. 5. They will receive the bottled water for up to 12 months. Pipe producers will pay for it and its delivery.
The industry also has offered to provide Doniphan County new PVC pipe that won't leach vinyl chloride. The new pipe will cost at least $150,000, Eckstein said. Testing should also begin soon to pinpoint which pipes need to be replaced.
Morby said his agency doesn't want to see people exposed to vinyl chloride above the federal limit -- even if it's not technically a violation of federal regulations.
"We know we've got levels above the (federal) limit in Doniphan County," Morby said. "We think the right thing to do is to make sure the community knows what's going on and to find ways to address the problem short-term and long-term.
"And that's what we're trying to do."