The city of Columbia’s 2016 water quality report praises its drinking water as being "naturally of very high quality and free of harmful chemicals and bacteria."
But after water leaves a treatment plant, it is possible it can pick up toxic impurities from lead service lines, which distribute water from water mains into homes, or from lead solder and brass fixtures that contain lead. That’s why the Environmental Protection Agency requires water systems to regularly monitor tap water for lead.
Under the agency’s Lead and Copper Rule, water utilities are supposed to test water from homes most likely to leach lead. However, a Missourian investigation of where Columbia tests its water found that the city does not prioritize homes at greatest risk for lead in the water during testing.
Instead, of the 180 samples the city has taken for lead and copper testing since 2007:
• Less than 7 percent of water samples came from single family homes with lead service lines or "highest risk" lead solder.
• Only six sites designated by the city as highest-risk for leaching lead were tested in 2016; one in 2013, four in 2010 and none in 2007.
• About 40 percent of water samples came from homes and buildings without any lead plumbing.
City officials insist their choice of sampling sites is appropriate because the goal of lead and copper testing isn't to identify highest risk sites, said Deidra McClendon, who took over Columbia’s lead and copper testing in 2016. "We’re looking at what our water does to lead.”
But to properly assess whether water chemistry is affecting plumbing, the city must test water from sites most likely to leach lead, said Marc Edwards, a professor of civil and environmental engineering at Virginia Tech whose lab exposed the water scandal in Flint, Michigan.
Otherwise, there’s no point. "If you aren’t testing the worst-case homes," Edwards said, "just throw the data into the garbage and tell the public they’re on their own."
A National Hazard
The water crisis in Flint, Michigan, which exposed residents to foul, discolored water with dangerously high levels of lead, eluded detection despite regulatory monitoring in part because water officials did not inspect water from the worst-case homes.
Independent analyses by Edwards’ lab later showed that nearly 17 percent of the tap water samples submitted by concerned residents exceeded the regulatory limit —which should have prompted instant remedial action. By then, residents had been cooking with and drinking the toxic water for months under the water department’s assurances that it was safe.
And it’s not just Flint. Lead in drinking water is a national problem. The Philadelphia Water Department is facing a class action lawsuit over questionable lead and copper testing practices. A Guardian investigation found that at least 33 U.S. cities, Philadelphia included, used sampling methods to “game the system” and minimize lead in their test results, such as pre-flushing the line, removing aerators, or "gently" filling the sampling bottle with tap water. Since 2012, approximately 2,000 water systems in the U.S. have failed compliance testing for lead in drinking water, according to USA Today.
Lead, despite its long association with health problems ranging from irreversible brain damage in children to infertility and miscarriage in adults, wasn’t banned in the U.S. as a plumbing material until 1986. By then, the material had become thoroughly embedded in the waterworks of many American cities in the Northeast and Midwest, primarily in the form of lead service lines.
When exposed to water, lead oxidizes and forms rust that collects along the interiors of pipes. An anti-corrosive additive called orthophosphate binds lead rust and keeps it in place. But changes in water chemistry, corrosion control or physical disruptions to the pipe can cause bits of lead-laden rust to flake off and enter the drinking water.
If excess lead levels are detected during testing, regulations require water systems to optimize corrosion control and alert the public so they can protect themselves by drinking bottled water or installing a filter for lead.
If corrosion control fails, utilities must begin replacing their lead service lines. However, when lead plumbing lies on the water customer’s side of the meter, responsibility falls upon the homeowners to replace their pipes, which can cost anywhere from $2,000 to $10,000.
The numbers of uncertainty
Despite some high test results in the past, Columbia has never violated the Lead and Copper Rule.
Water systems meet federal regulations if no more than 10 percent of their samples contain lead over a regulatory limit of 15 parts per billion, but health officials stress this is a technical standard. It's not based in public health because experts agree there is no safe maximum level of exposure for lead.
Since Columbia started official lead and copper testing, four water samples have exceeded the 15 parts per billion cutoff, including a record high of 36 parts per billion in 2013 from a north Columbia residence. Low amounts of lead have been detected in about 30 percent of the city’s official tap water samples taken between 2007 and 2016.
Independent testing by Columbia Public Schools in 38 of its buildings found lead in almost half of the 1,429 taps sampled over the summer, according to previous Missourian reporting. Forty-seven taps exceeded 15 parts per billion of lead, prompting the school system to post warning signs on the tainted faucets. The highest concentration of lead came from a classroom sink at Jefferson Middle School reading 2,100 parts per billion.
Because Columbia is a large water system on reduced testing, the Lead and Copper Rule requires a city of Columbia’s size to sample tap water from 50 sites every three years. The 90th percentile criterion allows five or fewer of the city’s 50 water samples to contain lead over 15 parts per billion. That means a handful of homes can have high lead levels and the system as a whole would still be in compliance.
The Lead and Copper Rule prioritizes the sampling of drinking water from locations by tier:
• Tier 1 sites — the highest priority — are single-family homes connected to lead service lines, use lead pipes or have copper pipes with lead solder installed after 1982.
• Tier 2 sites — the second priority — are buildings, including multi-family homes, that otherwise have the same criteria as Tier 1 sites.
• Tier 3 sites — the third priority — are single-family homes with lead solder installed before 1983.
Ideally, at least half the water samples should come from homes with lead service lines, while the remaining samples should come from houses with lead solder. Water systems should test as many Tier 1 houses as possible before moving on to Tier 2 and Tier 3 homes. Only water systems serving communities with an insufficient number of homes in the top three tiers can test sites without lead plumbing.
In 2016, Columbia took water samples from six sites labeled as Tier 1 by the city, including the office sink at the Columbia municipal power plant. But, as a non-residential building the power plant does not qualify as a Tier 1 site under the federal rule.
Before that, Columbia tested one Tier 1 site in 2013, four in 2010 and none at all in 2007.
A little over half of all samples taken by the city between 2007 and 2016 were labeled as Tier 3 sites by the city. But some of those sites included businesses, fire stations and schools, which, as non-residential sites, do not qualify as Tier 3 sites under the Lead and Copper Rule.
Approximately 40 percent of Columbia’s water testing samples during this time period came from buildings that don’t use lead plumbing at all, according to city records. The city also sampled tap water a total of 36 times from 16 houses built after the 1986 lead plumbing ban.
The city’s current testing regime has other problems. For example, the city has used its own employees to test water in their homes, which raises the question of a conflict of interest.
"Realize they’re sampling 50 homes once every three years to generate the information necessary to protect the entire city," Edwards said. "The only way that works is if you sample the worst houses."
Otherwise, "you’re providing a false sense of security," he said.
Lack of records
During Columbia's capital investment project in the late 1970s, there was a move to replace all the public lead service lines in use at the time, city water distribution manager Floyd Turner said.
"When I started 30 years ago," Turner said, "most of the lead [service] pipes were already removed. We’d run across one every now and then, and it was a kind of rule of thumb that if we found one, we’d replace them even if they weren’t leaking."
The city is not responsible for replacing private lead service lines on residential property, which run between the meter and the house, he said.
Nevertheless, water systems are supposed to keep an inventory of lead service lines, public or private, for testing purposes.
When the Lead and Copper Rule went into effect in 1991, water utilities had to survey the types of plumbing materials used in their distribution systems and inside residential buildings to find and maintain a pool of Tier 1 sites they could draw upon for testing.
Records submitted by the city to the Missouri Department of Natural Resources in 1992 showed that Columbia Water & Light had identified more than 320 possible candidate sampling sites, including 28 with lead service lines, though only eight agreed to participate at the time.
In the first round of testing, Columbia Water and Light drew water samples from eight homes with lead service lines and 52 homes with lead solder. All 60 sampling sites were deemed Tier 1 by the city.
Today, Columbia does not test the water from any of the initial 60 locations. Nor does the city know what happened to 20 of the 28 lead service lines it had identified in 1992 or where they are, even though the Lead and Copper Rule requires water systems maintain their inventory and update it when pipes are replaced.
After the Missourian gave Columbia Water and Light copies of the 1992 records the city had submitted to the Missouri Department of Natural Resources containing the locations of the eight lead services, Columbia Water and Light sent crews on site to dig up the pipes and found that they had been replaced with copper.
Deidra McClendon, who is in charge of lead and copper sampling for the city, relies on a dog-eared, coffee-stained manila folder less than an inch thick that holds all of Columbia’s historic lead and copper testing information, some of which she printed from the DNR website.
When she took over testing in 2016, McClendon said she knew nothing about the plumbing at the sampling sites chosen by her predecessor, or the criteria by which they were selected.
Some of the sites had been assigned the wrong tier, something McClendon didn’t discover until she sent an information-gathering materials survey to residents along with the sampling bottle for the 2016 round of testing.
"You and I assume there are records on everything," she said, "but probably some of those records don’t exist anymore."
McClendon said that she was also instructed by DNR to test sites that were representative of Columbia’s plumbing, rather than only those at highest risk. That’s why she included sampling sites such as homes without lead plumbing, businesses, schools and firehouses.
"We’re not required to have all 50 sites be Tier 1," McClendon said.
The Missouri DNR did not respond to multiple attempts by the Missourian to confirm her statement.
The federal rule requires Tier 1, 2 and 3 sites to be "family structures" or "multi-family residences."
City officials cited low participation rates as another reason why the city samples from so few high priority sites.
However, in Columbia there are plenty of homes that could meet lead and copper testing requirements.
According to records from the Boone County Assessor’s Office, about 900 houses built after 1982 but before 1986 might fit the criteria for Tier 1. More than 12,800 residences in Columbia were built before 1986. Since lead plumbing wasn’t banned until June of 1986, any of these houses could be potential Tier 1, 2 or 3 sites.
Nevertheless, said McClendon, "I felt like we did the best we could. I think (our testing) is very accurate and very representative of Columbia as a whole."
Yanna Lambrinidou, a former member of the Environmental Protection Agency National Drinking Water Advisory Council’s lead and copper rule working group and an adjunct professor at Virginia Tech, disagrees.
"The city has had over 20 years to identify Tier 1 homes," Lambrinidou said. "That they’re not sure what’s where suggests they’ve been ignoring lead and copper monitor requirements."
The Lead and Copper Rule is meant "to act as a canary in the coal mine" for the entire water system, Lambrinidou said.
That's because problems with lead in drinking water don’t just plague highest-risk homes, said Lambrinidou, who lived through and researched Washington’s lead drinking water contamination in the early 2000s.Many D.C. homes tested positive for lead even though they didn’t use lead service lines. Until 2011, pipes with up to 8 percent lead were considered "lead-free."
Just because a water system is in compliance with federal rules does not mean that a consumer’s drinking water is safe from lead, especially if the system isn't testing the highest risk sites, Lambrinidou said. "You could argue it’s negligence to say otherwise."
Francisco Vara-Orta contributed to this report.