Standing on a catwalk over a pool of water near the banks of the Ohio River, Frank Blaskovich points at a series of pipes draining into the far end of the pool.
"There, that's the river water coming in," says Blaskovich, water treatment manager for Wheeling, West Virginia.
His job is to take water from the Ohio River and make it into safe drinking water for his city of 30,000. But since 2008, the Ohio has been too salty, so he's had to dilute it with groundwater from backup wells. Blaskovich doesn't like doing this, because each added step costs money.
"The price of water will eventually go up, which probably will lead to a possible rate hike," he says. But he's blending the river water anyway because it's got high levels of bromide. Bromide is a salt, which by itself is harmless. But combined with chlorine at a drinking water plant like this one, it forms chemicals called trihalomethanes. Long term exposure to trihalomethanes increases the risk of bladder and other cancers.
Because of high bromide levels in the rivers, Wheeling and dozens of plants in Western Pennsylvania and West Virginia have violated the EPA's limits on trihalomethanes over the last three years. Over that time, 33 different Western Pennsylania drinking water systems have exceeded EPA standards for the carcinogen.
Bromides come from many places, including sea water, coal-fired power plants, and chemicals. But the Ohio's spike in bromide occured three years ago, and Blaskovich thinks that's no coincidence.
"That's when deep drilling for gas sort of took off in this area of the country," he says.
Each Marcellus shale gas well produces millions of gallons of salty water. The water is full of bromides, and until recently, drillers in Western Pennsylvania trucked this discharge to wastewater plants for disposal.
The plants could treat the water for metals and other pollutants, but not bromides. That requires expensive new technology. The plants would simply release the treated water — bromides and all — into rivers and streams.
But after trihalomethane levels started creeping up at drinking water plants, regulators took note. In March, the EPA expressed concern over Pennsylvania's handling of Marcellus discharge, and a month later, the state's Department of Environmental Protection asked drillers to stop sending wastewater to treatment plants.
DEP secretary Mike Krancer says a voluntary program would be quicker than making a new rule. "The industry — and I knew they would — did the responsible thing and complied, so we had compliance in 28 hours instead of 28 months," Krancer says.
According to DEP records reviewed by The Allegheny Front, the request stopped most, but not all, drillers from sending Marcellus shale brine to these plants. After the request, some facilities, like the Franklin Brine Treatment plant south of Erie, saw their oil and gas wastewater shipments drop by 70 percent.
Drillers say that they are recycling more water now, or sending it to Ohio, where it's injected into deep storage wells.
If drillers are sending much less of their salty water to treatment plants, bromide levels in the rivers should be going down. But, at least this year, that hasn't been the case, says Jeanne VanBriesen, a Carnegie Mellon scientist who's monitored bromide on the Monongahela River for the past two years.
"We thought in such a wet year, we would see almost no bromide, it would be below our detection limit in most of our samples, and it was not," she said.
But the question remains, where is the bromide coming from? Answering this question has been a mission for Stanley States, director of water treatment at Pittsburgh Water and Sewer Authority. This mission led him to the banks of the Clarion River, in Elk County, on a chilly November day. States has been measuring bromide in the source water for his Allegheny River plant in Pittsburgh.
Wearing black and yellow waders, he got into fast-flowing river. Into the current, he hurled a wire basket with a plastic jug inside. He'll analyze the contents of the jug at two separate labs to see if bromides from the Clarion could affect drinking water in the Allegheny further downstream.
"Source-water protection is part of what we're supposed to do as drinking water people," says States, during a break in his sampling. "Our treatment doesn't begin at the plant — it begins in the river system."
States has been working with University of Pittsburgh researchers to identify bromide hot spots. They've sampled above and below places like municipal and commercial treatement plants that have taken Marcellus Brine in the past.
Back in his office, he says that coal-fired power plants can raise bromide levels. But he found these aren't the biggest bromide producers.
"It's pretty clear the only places where there's significant increases in bromides are downstream from these industrial waste water plants," States says.
Citing scientific convention, States did not name the treatment plants he's monitoring, though he does provide their locations. This raised doubts about the study's credibility, for DEP Secretary Krancer.
"I thought that the charge being made particularly suspicious and baseless, certainly, because supposed names of violators were not disclosed. I have to look at that with a bit of suspicion," Krancer says.
The Allegheny Front shared the study with a handful of water scientists. They generally agreed that States used standard methods for a study of this kind. They said that it provides a snapshot of pollution levels in certain spots, not a definitive picture of the river's overall health. One strength, they said, is that the study covers a long period of time. States has sampled 38 places once a month for a year.
States stands by his study.
Krancer, the DEP secretary, says that it's probably too early to declare an ongoing bromide problem in the rivers.
Assuming the data are accurate, and that bromide levels are still high, does it mean that Marcellus brine is still the cause? That is unclear, says Carnegie Mellon's VanBriesen. She says that there could be other potential sources of bromide. These include wastewater from conventional oil and gas drilling, the kind that's been produced in Pennsylvania for decades.
"Almost anything to do with fossil fuel production has the potential to contain amounts of bromide we should manage and think about," VanBriesen says.
So does this mean that water from hydraulic fracturing wasn't really a problem in the first place? That's what some plant operators say. One of these operators is Paul Hart, who owns three oil and gas treatment plants, including the Franklin Brine plant. Hart said that DEP used "bad science" in asking drillers to steer clear of his plant.
But VanBriesen, of Carnegie Mellon, disagrees.
"We know this wastewater has bromide in it," she says. "Any waste you take out of the system that contains bromide will reduce amount of bromide in the basin."
In the spring, the EPA will tighten their standards for trihalomethanes. That has water plant operators like States worried. "And the public will be very upset if they have to receive public notification that their water doesn't comply with the safe drinking water act," States says. "They're going to want answers."
Chief among their questions, States says, is who's responsible for getting bromides out of the rivers.
Bromide in the Allegheny River and THMS in Pittsburgh Drinking Water: A Link with Marcellus Shale Drilling
Stanley State's survey of the Allegheny River and its tributaries. The study includes bromide data from above and below industrial treatment plants, coal-fired power stations, and municipal treatment plants. It was presented at an American Water Works Association conference last month.
Western Pennsylvania drinking water systems that have exceeded the EPA's limit of 80 parts per billion annual average for trihalomethanes, since 2008. Source: Pennsylvania Department of Environmental Protection, Drinking water reporting system.
The Allegheny Front's Ann Murray also reported on this story.