This 1940 photo shows Northwestern Smelting and Refining at 2523 Hiawatha Ave. in Minneapolis. State regulators told the EPA in 2002 they found no information that a smelter once operated at the site.
By Alison Young, USA TODAY
Ken Shefton is furious about what the government knew eight years ago and never told him - that the neighborhood where his five sons have been playing is contaminated with lead.
Their Cleveland home is a few blocks from a long-forgotten factory that spewed toxic lead dust for about 30 years.
The Environmental Protection Agency and state regulators clearly knew of the danger. They tested soil throughout the neighborhood and documented hazardous levels of contamination. They never did a cleanup. They didn't warn people living nearby that the tainted soil endangers their children.
INTERACTIVE: Explore more than 230 lead-factory sites
PHOTOS: Historical photos of forgotten lead factories
VIDEO: Ghost Factories: A failure to protect the public
"I needed to know that," Shefton said. "I've got a couple of kids that don't like to do nothing but roll around in the dirt."
More than a decade ago, government regulators received specific warnings that the soil in hundreds of U.S. neighborhoods might be contaminated with dangerous levels of lead from factories operating in the 1930s to 1960s, including the smelter near Shefton's house, Tyroler Metals, which closed around 1957.
Despite warnings, federal and state officials repeatedly failed to find out just how bad the problems were. A 14-month USA TODAY investigation has found that the EPA and state regulators left thousands of families and children in harm's way, doing little to assess the danger around many of the more than 400 potential lead smelter locations on a list compiled by a researcher from old industry directories and given to the EPA in 2001.
In some cases, government officials failed to order cleanups when inspectors detected hazardous amounts of lead in local neighborhoods. People who live nearby - sometimes directly on top of - old smelters were not warned, left unaware in many cases of the factories' existence and the dangers that remain. Instead, they bought and sold homes and let their children play in contaminated yards.
PHOTOS: In the fallout zone: A Cleveland neighborhood
The USA TODAY investigation shows widespread government failures taking several forms:
•A failure to look. At dozens of sites, government officials performed cursory inquiries at best. In Minnesota, Indiana and Washington, state regulators told the EPA they could find no evidence that some smelters ever existed.
Yet in those states and others, reporters found the factories clearly documented in old insurance maps, town council minutes, city directories and telephone books - even in historical photos posted on the Web.
•A failure to act. In Pennsylvania, Maryland and Wisconsin, the EPA sent investigators to scores of sites from 2004 to 2006 after verifying a lead smelter once operated. The investigators recommended soil tests in the neighborhoods. Most of the tests were not done.
•A failure to protect. Even when state and federal regulators tested soil and found high levels of lead, as they did around sites in Philadelphia, Cleveland, Chicago and Portland, Ore., they failed for years to alert neighbors or order cleanups. Some kids who played in yards with heavily contaminated soil have dangerous levels of lead in their bodies, according to medical records obtained by USA TODAY.
In response to the investigation and USA TODAY's soil tests in 21 neighborhoods, government officials are taking action at old smelter sites in 14 states, ranging from reopening flawed investigations to testing soil to cleaning up contaminated property. In March, New York City officials closed four ball fields in a Brooklyn park after learning from USA TODAY that the area was a former smelter site with elevated levels of lead.
"EPA and our state and local partners have overseen thousands of cleanups, through a variety of programs," said Mathy Stanislaus, an EPA assistant administrator. "Unfortunately, some of the sites USA TODAY identified have not yet been addressed or investigated by EPA. EPA will review USA TODAY's information to determine what steps can be taken to ensure Americans are not being exposed to dangerous levels of lead."
The EPA says it has worked with states to assess most of the sites on the 2001 list but that record-keeping is "incomplete" for many. Eighteen sites received some kind of cleanup but most weren't considered dangerous enough to qualify for federal action.
"I am convinced we have addressed the highest-risk sites," said Elizabeth Southerland, director of assessment and remediation for the EPA's Superfund program. "Absolutely and positively, we are open to reassessing sites that we now feel, based on your information, need another look."
EPA staff members said additional site reviews are underway, including checks of 48 sites the agency determined were never assessed. And the EPA said it will work with Ohio environmental regulators to re-examine the Cleveland neighborhood near Shefton's home to see whether a cleanup evaluation there is appropriate.
Ken Shefton and his family aren't waiting for the government to do a cleanup. His 6-year-old son, Jonathan, was diagnosed this spring with having an elevated level of lead in his body, Shefton said: "That was the last straw." He's in the process of selling his home. The family moved to another neighborhood last week. "Somebody needs to take care of this problem, or inform the people in this neighborhood," he said.
Concerns surfaced a decade ago
Most of the nation's lead factories - some huge manufacturing complexes and others tiny storefront melting shops - had been largely shuttered by the 1970s and 1980s. Often known as smelters, they emitted thousands of pounds of lead and other toxic metal particles into the air as they melted down batteries and other products containing lead.
The particles would land on nearby properties, potentially mixing with lead dust from automobile exhaust or paint chips - significant sources, says the government - to create a hazard. Children who play in lead-contaminated soil, sticking dust-covered hands or toys in their mouths, over time can suffer lost intelligence and other irreversible health problems.
In April 2001, environmental scientist William Eckel published a research article in the American Journal of Public Health warning about the dangers of old smelting factories. While working on his Ph.D. dissertation, Eckel had identified a historical smelting site unknown to federal and state regulators and wondered how many other sites had been forgotten over time, their buildings demolished or absorbed by other businesses.
Eckel used old industry directories, which he cross-referenced with EPA databases, to come up with a list of more than 400 potential lead-smelting sites that appeared to be unknown to federal regulators.
Eckel confirmed that 20 of the sites' addresses were factories - and not just business offices - using Sanborn fire insurance maps, which detail the historical uses of individual pieces of property. An additional 86 sites were specifically listed in directories as "plant" locations. He paid to have soil samples tested from three sites in Baltimore and five in Philadelphia. All but one of the samples exceeded the EPA's residential hazard level for lead in areas where children play.
Eckel's article warned that the findings "should create some sense of urgency for the investigation of the other sites identified here because they may represent a significant source of exposure to lead in their local environments." The research indicates "a significant fraction" of the forgotten sites will require cleanups - likely at state and federal expense - because most of the companies went out of business long ago.
Buried by bureaucracy?
Eckel's research caught the attention of the EPA, which in 2001 asked him for a copy of his unpublished list, then shared it with EPA regional offices.
Records obtained under the Freedom of Information Act offer few details of the exact instructions the EPA gave to those receiving the list. Southerland, the EPA Superfund official, said the agency didn't provide regional offices any additional money or people to evaluate the old smelter locations. It asked only that the sites be put in their queues for possible assessment.
"We only have about 80 people and $20 million each year to do our site assessment program," Southerland said. About half of that money is sent by the EPA to state agencies.
Cleaning up contamination left by a smelter can be expensive. In Omaha, the EPA has cleaned up 10,000 residential yards and spent nearly $250 million addressing a former smelter there that wasn't on Eckel's list because it was already known to the agency. Many of the factories on Eckel's list were smaller operations.
With limited resources and many contaminated sites, state and federal environmental officials have to prioritize assessing sites they consider of greatest risk, Southerland said, and drinking-water contamination tends to trump soil contamination.
In addition, Southerland said, the EPA is authorized to clean up contamination only if it can show it came from an industrial release. That can be tricky to determine in some urban areas, where the agency says it's not uncommon to find high levels of lead contamination in soil, "particularly in large cities ... due to historic gasoline emissions from vehicles, aerial deposition from industrial facilities, and lead paint," the EPA said in a statement.
The government's efforts to investigate the sites on Eckel's list varied widely, records show. Dozens were never investigated. Others received a cursory records review or a "windshield survey" - a drive-by type of visit. Soil was tested at some sites, but the testing in some cases was limited to the former smelter's property boundaries and ignored where the wind might have carried airborne contamination; in other cases, testing was also done in nearby neighborhoods.
By 2005, concerned the list of 464 sites had been too large of a workload for the regions, officials at EPA headquarters launched their own assessment effort, Southerland said. The focus was on having regions examine a sampling of 31 sites from Eckel's list. They concluded many lacked evidence that they were ever smelters, according to a 2007 report obtained under FOIA marked "For Internal EPA Use Only." The report said only one of the sites determined to have been factories, Loewenthal Metals in Chicago, might qualify for a federal cleanup and the rest were being addressed by state regulators. Southerland said a North Carolina site ultimately received a federal cleanup.
Only six of EPA's 10 regional offices had undertaken some sort of smelter discovery initiative, according to the 2007 internal EPA report. Two of those initiatives - one by federal officials in Pennsylvania and Maryland, the other by EPA Region 5 and Michigan state officials - focused on sites from Eckel's list, the report said.
Michigan regulators took actions at some Detroit smelters after the Detroit Free Press in 2003 did historical research into 16 Detroit sites on Eckel's list and found smelting or foundry work at most of them. Only one site was being cleaned up at the time of the report. In 2006-07, cleanups occurred in two more neighborhoods, according to a state contractor's report.
But in scores of other cases, USA TODAY found government agencies didn't do much to protect families and children - even when their own tests showed dangerous levels of lead where people live.
Reporters scour 464 sites
The USA TODAY investigation set out to determine which sites remained unaddressed and to examine the depth and quality of any government assessments.
Reporters researched all 464 sites in 31 states that were on Eckel's list to determine how many were factories, rather than just business offices - and what, if anything, had been done to clean up those hazardous enough to threaten people living nearby.
Reporters spent weeks in the basement of the Library of Congress, researching its extensive collection of Sanborn maps. Maps showing smelting or factories were located for more than 160 sites - including many that regulators never looked for because they lacked exact street addresses. Reporters researched old phone books and city directories, archival photograph collections, old business directories, property records and corporation filings. They filed more than 140 federal, state and local public records requests with environmental, health and other government agencies to determine what, if any, assessments had been done of the sites and the risks posed to people nearby.
As a result, the investigation found evidence of smelting, foundries or lead manufacturing at more than 230 sites in 25 states on the list of forgotten factories.
The failure to protect
Ken Shefton, his wife and five boys lived until last week in a Cleveland neighborhood a few blocks northeast of the former site of the Tyroler Metals smelter. The area's two-story wood homes, mainly built around 1900, are flanked by factories, both operating and abandoned. A smelter operated at the Tyroler site from about 1927 through 1957, according to the state's report. Smelting no longer occurs at the site, which is now a scrap yard with a different owner.
In 2002 and 2003, state regulators from the Ohio Environmental Protection Agency- working at the request of the federal EPA - tested 12 samples of soil around the old site and in the nearby neighborhood. All but one showed lead contamination above the EPA's residential hazard level of 400 parts per million (ppm) of lead in bare soil where children play. Nine of the samples had lead levels ranging from twice to five times the hazard level, according to the state's report.
The results indicated a possible "airborne depositional pattern or plume towards the northeast," the report said. In layman's terms: a fallout zone.
The state's research also identified that other smelters had been on adjacent properties dating to 1912, as well as a currently operating lead-manufacturing plant nearby. "A problem interfering with future investigation is attribution of lead contamination, due to multiple sources," the state's report said.
No matter the source, regulators never warned residents about what they found, and no cleanup occurred.
State regulators at the Ohio EPA said that without a specific polluter to blame - and force to pay for cleanup costs - there was nothing more they could do. "There are no Ohio EPA monies set aside and dedicated for this type of cleanup," the agency said in written responses to questions. "Our enforcement program focuses on responsible parties with the authority to legally compel them to fund cleanup."
Still, state regulators said that more than seven years ago they "recognized there could be potential for a health concern based on the sampling results." They said they fulfilled their duty by putting their findings about the neighborhood in a report and sending it to the EPA's regional office in Chicago. The state says it sent the report about Tyroler Metals, along with reports on eight other historical Cleveland smelter sites, to the director of the Cleveland Department of Public Health in June 2004.
Either agency could have followed up, the state said. Neither did.
Officials at the EPA regional office said that because the site didn't meet criteria for federal Superfund action, it was the state's responsibility. Federal and state officials now plan to review the site to see whether a cleanup evaluation is appropriate, the EPA said in a written statement.
Current and former Cleveland health department officials - including Matt Carroll, who at the time was health director, and Wayne Slota, who at the time was in charge of the lead poisoning prevention division - said they don't remember receiving the state's letter and reports about Tyroler Metals.
The only smelter issue they remember involved a different site on Eckel's list: Atlas Metals, where a city park had been built atop the old smelter site and state investigators had observed children playing in dirt that tests showed was significantly contaminated.
Of the 17 Ohio sites on Eckel's list - in Cleveland, Cincinnati, Columbus and Toledo - Atlas Metals was the only one records indicate received a cleanup.
A neighborhood suffers
"I'm concerned. I really don't know what to do," said McKinley Woodby, as he held his then-15-month-old son, Damien, on his lap. "I'm just a renter. I'm on a fixed income, so it ain't like I can dig the front yard up and bring in new dirt."
"I'm not going to let (Damien) back in the yard, I know that," he said, sitting on the front steps of their home about four blocks from the Tyroler Metals site.
When USA TODAY tested soil in the family's yard where Damien played, the results showed potentially dangerous contamination in four of five samples, ranging from 577 to 1,035 ppm. Although the EPA uses 400 ppm as its residential hazard level, California's environmental health agency has set 80 ppm as the level it says will protect children who regularly play in the dirt from losing up to 1 IQ point over time.
Damien's blood was checked a few weeks before USA TODAY tested the yard. Health department records show he had a blood-lead level of 4. That's below the federal action level - set in 1991 - but current science indicates children with levels below 5 are at risk of having decreased academic achievement.
Blood test results filed with the Ohio Department of Health show that during 2007 through mid-2011 in the smelter's ZIP code about 350 kids under age 6 had reported blood-lead levels of 5 or higher. About the same number had blood-lead levels of 2 to 4. There is not a definitive way to know how prevalent lead poisoning is in the area because not all children are screened and some tests are less accurate than others.
How much the lead in the dirt is contributing to the children's blood-lead levels is unclear. But experts say that soil is an important component, along with deteriorating lead-based paint in older homes and contaminated house dust.
Bruce Lanphear, a leading expert on childhood lead poisoning, said his research has estimated that for the average child about 30% of the lead in the body comes from contaminated soil, about 30% from contaminated house dust - which includes particles of flaking paint - and about 20% from water.
"Those were the major sources, so they're all fairly important," said Lanphear, a professor of children's environmental health at Simon Fraser University in British Columbia.
A child's lead exposure can be very individualized, he said, depending on geography. For some children, it might be all about paint. "If you were to look at a community that's adjacent to a smelter, it might be that it's 80% soil, or 90% soil."
'Oh, my gosh, no, I didn't know'
In Chicago, officials have known for years about a neighborhood where contamination could pose a danger and have done little to address it. Walsh Elementary School in Pilsen is just down the block from the former site of Loewenthal Metals.
Delinda Collier said she had no idea the site used to be a lead smelter and was contaminated. There were no warning signs on the property. "Oh, my gosh, no, I didn't know," said Collier, 38, who rents an apartment across the street and lets her dog play on the vacant lot. "I'll bet nobody else does either."
Federal and state regulators knew.
Tests by the state in 2006 found the former smelter's vacant lot contaminated with up to 5,900 ppm of lead - more than 14 times the amount the EPA considers potentially hazardous in areas where children play.
"Since this site is in a residential area, the possibility of exposure is high," according to the report state officials sent to the EPA, which commissioned the work. But the site wasn't bad enough to qualify for its Superfund list, and the report was archived.
State regulators at the Illinois EPA said Loewenthal Metals was one of about 50 old smelter sites in Chicago they reviewed to varying degrees at the request of the U.S. EPA. The Loewenthal site had even been highlighted in the 2007 EPA headquarters report as the only site examined under its smelter initiative that might need a Superfund removal action.
Still, it fell through the cracks.
"We never got any follow-up instructions from them on what additional things to do with the reports we sent up to them," said Gary King, who was manager of the state agency's division of remediation management until he retired in December.
"Nonetheless, as a result, frankly, of the (open records) request that came in from USA TODAY and going back in and looking at this information ... we concluded that it would be best to send in what we call a 'removal action' referral," King said. That means the state is now formally asking the EPA to remove the contamination from the property.
The state also is formally asking the EPA to clean up a second Chicago site, Lake Calumet Smelting, where its tests in 2004 found high levels of lead - up to 768,000 ppm - on the former factory's property. The nearest homes are about a half-mile away, records show.
The failure to act
Even when officials did identify factory sites and nearby neighborhoods that could be contaminated, they failed to follow through.
The EPA's Philadelphia regional office developed one of the agency's most comprehensive smelter initiatives in response to Eckel's report. Officials there sent contractors in 2005-06 to visit most of the 71 factory sites listed in Pennsylvania, Maryland and Virginia.
The assessments confirmed dozens of the sites had had smelters, reports show, with 34 of them in troubling proximity to homes, parks and schools. As a result, EPA contractors recommended soils nearby be tested. Despite the passage of years, testing has been done at 10 sites, fewer than a third, records show.
The EPA now says the site assessment process is ongoing and the agency must prioritize its use of resources. In some cases, the EPA may not agree with its contractor's recommendations. Still, the EPA said it plans an additional assessment at several sites in late 2012 or early 2013. The "lead smelter sites at this time do not seem to pose the same threats we are encountering at other sites in the region," the EPA said.
The threat seemed serious to others in 2004.
At that time, state and federal health officials distributed a health alert to doctors with a map of the Pennsylvania locations on Eckel's list. The alert by the Pennsylvania Department of Health and the federal Agency for Toxic Substances and Disease Registry recommended doctors consider doing blood tests on children living near the sites to look for lead poisoning.
The EPA's Philadelphia regional office, however, says it sees no need to put out general warnings to neighbors of old smelter sites. "This type of approach would unnecessarily alarm residents and community members," it said. The office also said it saw no need to tell Maryland's state environmental agency about the 11 smelter sites in its state on Eckel's list. Nor did the EPA region alert the state agency that federal contractors had recommended soil testing around five of them.
USA TODAY provided Maryland officials the locations of the sites - and copies of the EPA's reports.
The EPA's failure to share such information is unusual, said Art O'Connell, chief of the Maryland Department of the Environment's state Superfund program. "I don't know what happened in this particular case, but it's certainly not the norm," he said.
As a result of the information provided by the newspaper, O'Connell said, the state recently examined the sites and determined that two former factories in Baltimore warrant further investigation: Industrial Metal Melting and Dixie Metal Co. The state has asked the EPA for funding to do soil testing and other investigation at the sites this year.
As for the three other factory sites where EPA's contractors recommended tests, O'Connell said his department believes they were small operations and had little impact on soil.
The failure to look very hard
Philadelphia-based officials started investigations; other EPA regions did far less.
Of the 120 sites on Eckel's list in New York and New Jersey, the EPA office responsible for those states sent inspectors to 14 locations. (USA TODAY found historical fire insurance maps and other documents showing evidence of smelting at 53 sites in those states.)
And even though the entire focus of Eckel's list involved smelters that had closed long ago, the EPA in 2002-03 inexplicably sent inspectors looking for active smelters at only nine of the locations.
"On each occasion, upon reaching the site where the smelter was supposedly operating, the inspector found the smelter had been closed down long ago," said Philip Flax, an EPA senior enforcement team leader, in a letter to USA TODAY that accompanied copies of some inspection reports.
In 2005-06, the EPA visited four more sites in New York and one in New Jersey.
The New Jersey Department of Environmental Protection had files on only five of the 31 sites listed in its state, according to the department's responses to 31 separate open records requests it required USA TODAY to file. Only two of the files showed evidence the sites were smelters or lead factories. Yet USA TODAY later found evidence that 12 additional sites were factories. The state is now working with EPA to investigate, DEP spokesman Lawrence Hajna said. He also now says the department has located case files on some sites it told USA TODAY it didn't have.
In 2002 and 2003, the New York Department of Environmental Conservation did an "informal investigation" at some of the 89 sites listed in the state, spokeswoman Emily DeSantis said.
Four sites were known to the department and undergoing cleanups. At the remaining sites, the department concluded there was "no evidence" of environmental impacts or "no apparent impact," according to information provided by DeSantis.
Yet the department provided records documenting staff visits to just 13 of those sites. Others were assessed by the department's regional offices, DeSantis said, but the department had no record of those evaluations. There was no soil testing at any of the sites, she said, but USA TODAY's findings will be reviewed for possible follow-up.
In other states, USA TODAY repeatedly located smelters that regulators said their extensive research found no evidence had existed.
The Indiana Department of Environmental Management told the EPA in 2002 they could not find the site of the former Chas. Braman & Sons factory in the north-central Indiana town of Plymouth. The list provided to them by the EPA had only a post office box as an address. "Numerous historical industrial directories, as well as Sanborn maps, were consulted without finding any reference to the site," the state said in a 2002 report sent to the EPA.
The newspaper found a street address for the plant listed in a 1959 edition of Plymouth's telephone directory. A call to Plymouth's City Hall produced council minutes beginning in 1954 showing that emissions from the plant were a source of citizen complaints. According to a 1956 article from a local newspaper that Plymouth's city attorney found in the town's history museum, the Chas. Braman & Sons "smelting plant manufactured granular aluminum, solder and lead."
In response to USA TODAY's findings, state regulators sent staff to Plymouth and took 24 off-site soil samples from various locations near the former facility. Another six samples were taken on the factory site, which is now a granular aluminum company.
All the state's tests showed lead levels below federal guidelines; many did not detect any lead. "We did not see anything we were concerned about," said Mark Jaworski, a project manager in the state's site investigations section. The current owner of the aluminum company on the property, John Oliver Sr., said there has been no lead smelting since the Bramans' sold their factory around 1965.
Minnesota regulators told the EPA in a 2002 memo they were unable to confirm whether any of the seven sites in their state had been smelters. USA TODAY found evidence of historical smelting at two of them.
A state employee checked corporation records and did a drive-by of the former Hiawatha Avenue location of Northwestern Smelting & Refining in Minneapolis and noted a construction company and a bus line were among current businesses there. "No information available as to the operation of a smelter at this location," wrote Gary Krueger in his 2002 assessment.
The newspaper found photographs from the 1940s of the smelter in operation posted on the Minnesota Historical Society's website. A reporter located a historical Sanborn fire insurance map at the Library of Congress showing three smelters there at one time.
Krueger told the EPA in 2002 he couldn't find evidence of a National Lead smelter, which had been listed in St. Paul without a street address in old industry directories. "Additional use of state resources cannot be justified based solely on name of potential facility somewhere in St. Paul," says the state's report.
A reporter located the factory by searching through old indexes to Sanborn fire insurance maps. The map shows the National Lead plant was in a warehouse district near the Mississippi River and what is now Harriet Island Regional Park and describes it as a manufacturer of lead pipe, babbitt, solder and printers' metals; it also shows melting kettles.
After being given the photos and maps found by USA TODAY, Krueger recently visited the St. Paul site and made a second visit to the Minneapolis site. Krueger, a project manager in the state's Superfund program, noted the areas have undergone redevelopment.
"Quite honestly, it really doesn't change anything," he said. Without more proof of a danger, Krueger said, his department can't justify doing any soil sampling.
USA TODAY tested soil near the former National Lead site in St. Paul and found elevated levels in street-side public rights-of-way ranging up to 539 ppm. None of the three samples taken inside the park - which is in the river's flood plain - showed lead levels above 400 ppm, the EPA's hazard level for children's play areas. Near the Minneapolis smelter site, USA TODAY's tests found varying levels of lead.
•COMING NEXT: More tests, more contamination
Contributing: Peter Eisler, Adam Kerlin, Brad Heath, Nicole Dao, Paul Monies and Barbara Hansen.