Elevated levels of lead in Washington, D.C.'s drinking water became national news in 2004, reviving interest in a public health issue that had largely faded from the media spotlight since the U.S. Environmental Protection Agency finalized the Lead and Copper Rule (LCR) in 1991. Although most utilities continue to comply with the lead regulations without too much difficulty, others are discovering that changes made to comply with other regulatory requirements can affect lead levels in unforeseen ways.
The LCR is unique because it requires water providers to address contamination that often results from factors beyond their direct control. For example, lead is rarely present in water supplies, and drinking water usually contains little lead following treatment. However, lead in service lines, household plumbing fixtures, and solder can leach into drinking water under certain conditions. Because the corrosivity of water often plays a key role in this process, the LCR requires that utilities take steps to reduce corrosion within their distribution systems.
Drinking water providers are required to collect samples periodically from a certain number of residences or buildings considered at high risk of contamination. The LCR establishes a so-called action level of 0.015 mg/L of lead based on the 90th percentile level of tap water samples. If the results of its sampling exceed the action level, a utility may be required to conduct a host of measures, including additional monitoring, treatment to control corrosion, public education, and replacement of lead service lines.
Overall, the LCR has been a success, said Jim Taft, executive director of the Association of State Drinking Water Administrators in Washington, D.C. Lead levels present in drinking water have declined significantly compared to levels present before 1991, he said. However, he acknowledged that the LCR is “not a straightforward regulation.” Because lead in drinking water tends to leach from the distribution system, Taft said, adjustments made by a drinking water provider to comply with another regulation can inadvertently make the water more corrosive and cause more leaching to occur.
For more than a decade since Greenville Utilities in Greenville, N.C., began testing in 1992, its lead levels never came to close to exceeding the EPA's action level. However, that changed abruptly in 2004. “We were surprised,” said Barrett Lasater, Greenville's water plant manager, when the results of testing conducted that year indicated that 22% of samples contained lead in concentrations above the action level. Follow-up testing in May and June 2005 found that 27% of samples surpassed the action level.
Because none of the 565 miles of pipeline that comprise the utility's distribution system contains lead, Greenville is certain that the contaminant is leaching from faucets containing lead and lead-based solder used to connect pipes in the homes of its customers. Although the utility is still trying to determine why lead levels spiked in 2004, the presumption is that changes to the disinfection process at the utility's water treatment plant likely reduced the effectiveness of the corrosion inhibitor used to prevent lead from leaching from customers' faucets and pipes.
Between the time it tested for lead in 2001 and its subsequent testing three years later, Greenville made a series of changes to its drinking water treatment plant to comply with EPA's regulations regarding disinfection byproducts (DBPs). Considered by the EPA to be a potential health threat, DBPs form when organic matter in drinking water comes in contact with chlorine. In December 2002, Greenville began using chloramines, rather than free chlorine, to provide secondary disinfection, because chloramines do not create DBPs. Then in August 2003, the facility switched from chlorine to ozonation as its method of primary disinfection before filtration.
Pilot testing conducted by the utility before making the changes did not indicate that they might make the treated water more corrosive, Lasater said. “We had not anticipated any problems from the treatment changes,” he said. Other treatment plants in North Carolina had made similar changes without experiencing elevated lead levels. However, Greenville was using a polyphosphate corrosion inhibitor, while the other plants employed an orthophosphate inhibitor.
Greenville hired AH Environmental Consultants Inc., Newport News, Va., to study its corrosion control approach within its distribution system. Although the testing found that the water was not more corrosive than it had been before the changes to its disinfection system, the study determined that orthophosphate likely would provide the most effective corrosion control. Based on these findings, Greenville switched from polyphosphate to orthophosphate in 2004.
The most recent round of testing, conducted in November and December 2005, would appear to indicate that the change is having a positive effect, with only 16% of samples exceeding the action level. “We're encouraged that we're starting to see some improvement,” said Lasater. “But obviously we're not meeting EPA's requirements yet.”
In addition to providing free lead testing to its customers, Greenville is conducting research to discern the cause of the lead levels. The utility collected faucets and associated piping from six residences. Using a technique known as X-ray diffraction, researchers will study the pipes to determine the form of lead that is present in the plumbing and to analyze the scale that coats the pipes. The utility plans to collect faucets and piping from another six residences in the future to see whether the coating that forms on the pipes differs from the previous coating, Lasater said.
Another theory under examination by researchers holds that a change in the ratio of chloride to sulfate in the water may have caused the elevated lead levels, Lasater said. Unlike most treatment plants in North Carolina, Greenville uses polyaluminum chloride as a coagulant, instead of aluminum sulfate. “We're working on two different fronts to try to find out what the cause is,” said Lasater, and to determine if additional treatment changes should be made. “We're in new territory here,” he said.
LANSING'S LEAD LINES
Unlike Greenville, the source of lead in the water provided by the Board of Water and Light (BWL) in Lansing, Mich., is clear. There is “no question” that the approximately 12,500 lead service lines connecting some homes to the BWL's water main are the most likely source of lead, said John Strickler, BWL's communications director. Although sampling results have never exceeded EPA's action level, recently they have come “uncomfortably close,” said Strickler. Test results in 2002, the most recent year in which BWL tested for lead, indicated that lead levels had reached 0.013 mg/L at the 90th percentile.
In 2004, PWL began a $36 million program to remove all of its lead service lines by 2014.
Although the source of the lead is apparent, BWL is still trying to determine why lead levels have been increasing. In 1992, lead levels were slightly more than 0.01 mg/L at the 90th percentile, prompting BWLto begin adding a proprietary orthopolyphosphate blend as a means of reducing corrosivity in the distribution system, said Bill Maier, BWL's water quality administrator.
“We had very good results,” said Maier. In 2000, BWL switched to a different vendor that used an orthopolyphosphate blend with a slightly different mix of components. Around this time, BWL began to have “deteriorating results,” Maier said. Whether that was a key change in terms of lead levels in the water, Maier said, “we're still not absolutely sure.” The utility also reduced the orthophosphate residual slightly in response to requirements set by the Michigan Department of Environmental Quality regarding minimum phosphate levels.
BWL hired the consulting engineering firm Malcolm Pirnie, headquartered in White Plains, N.Y., to review the utility's efforts to control corrosion. The firm is developing recommendations that BWL can follow to stabilize the scale that forms naturally in the lead service lines, Maier said. Recommended changes are expected to include using additional orthophosphate for corrosion control and maintaining “slightly tighter” control over such water quality parameters as pH and alkalinity.MORE STUDY NEEDED
Because of the potential for treatment changes to affect lead levels in distribution system, utilities need to understand what comprises the scale that occurs in their pipes, said Michael Schock, a chemist in the EPA's Office of Research and Development. When pipes or lead service lines are removed from service, they can be analyzed by X-ray diffraction or other techniques to determine the nature of the scale in the system. In this way, Schock said, utilities have information to help predict what might become destabilized and enter the finished drinking water following certain kinds of treatment changes.
The complicated nature of the problem and the difficulties inherent in studying it mean that utilities are “learning the science after the fact,” said Marc Edwards, professor of civil and environmental engineering at Virginia Polytechnic Institute and State University. Because so much is unknown about how different treatment changes can exacerbate the leaching of lead, drinking water providers should plan after any treatment change to monitor their systems for increased lead levels beyond what the LCR requires, Edwards said.
Water providers typically rely on customers to perform the sampling, which introduces another variable that can complicate efforts to comply with the LCR, said Kathy Moriarty, water quality manager for the Bangor (Maine) Water District. “You have to have the cooperation of your customers to do this for you,” said Moriarty. “Sometimes [customers] can't or they don't want to” cooperate, she said.
To ensure that customers continue to cooperate with sampling efforts, the water department in Cedar Rapids, Iowa, samples more often than required by the LCR, said Shelli Grapp, the department's administrative affairs manager. Because its lead levels do not exceed the action level, the department is allowed to sample every three years. However, Cedar Rapids monitors annually to help retain its current pool of sample sites, Grapp said. “We're concerned that if we only contact people once every three years, we are going to lose continuity,” she said.
Moreover, if Cedar Rapids were to experience high lead levels, Grapp said, “we really don't want to wait three years down the road to determine that.”
— Landers is a freelance writer and editor based in Austin, Texas.