American Toxic

In a March 2007 affidavit, Rosenfeld explained that one extra cancer per million is the level considered acceptable by the EPA, and 100 extra cancers per million is a mandatory clean-up level. In Somerville, he wrote, the cumulative cancer risk to the community is 11,434 extra cancers per million—demanding an immediate remediation and possible evacuation of the area.

"That Somerville tie plant is the most egregious, horrible-acting facility...intentionally burning and poisoning the community...that I have ever seen in my life," said Rosenfeld, principal of Santa Monica, California-based Soil/Water/Air Protection Enterprise, in a December 2006 deposition. "There was a lot of ash and soot from the boiler that basically coated the entire community...and that contamination is still blowing around Somerville today."

Phillip Goad, a toxicologist who conducted environmental testing in Somerville for BNSF Railway, dismissed Rosenfeld's methodology in an August 2007 deposition, arguing that attic dust is "not relevant" for determining risk exposure since residents generally do not spend time in attics.

Goad collected carpet dust in nine houses, including the Fausts', and reportedly found no evidence of contamination that could be linked to the tie plant. He called the carpet a "historical reservoir" that residents contact on a daily basis. When pressed, however, Goad said that he never asked how frequently the families vacuumed or shampooed the carpets—which, he admitted, "would be a factor."

Rosenfeld, meanwhile, called attic dust a "time capsule" for environmental contamination. He argued in his deposition that it represents the best way to quantify historic exposure. In fact, he said, his estimates are conservative since the contaminants in the dust had been diluted over time.

Responding to concerns from Somerville residents, the Texas Department of Health Services conducted three cancer cluster studies in Burleson County from 2004 to 2006, finding a normal-to-expected rate of cancer among residents—a conclusion used by BNSF attorneys to bolster their case.

But the state's cancer cluster studies are flawed, says toxicologist James Dahlgren, since they rely on health information collected from death certificates, which are often inaccurate.

More significantly, Somerville has no hospital. As a result, Dahlgren says, the studies excluded many residents and tie plant employees who moved away or died in hospitals outside Burleson County.

These limitations are noted in the state's final reports: "Cancer incidence data are based on residence at the time of diagnosis. It is possible that some residents who may have been exposed and developed cancer no longer lived in the area at the time of diagnosis so were not included in the data."

James Dahlgren has spent the last year overseeing the first-ever epidemiological study for the entire town of Somerville. Nearly every weekend this summer, a team of his UCLA graduate students flew 1,500 miles to Houston, then commuted by rental car to Somerville, where they went door-to-door asking residents to complete extensive health surveys. So far, they have collected 600—half their goal.

Many residents have also undergone physical examinations, providing blood and urine samples for analysis. Dahlgren will collect and sort all the information to determine whether Somerville residents have a significantly higher incidence of cancer, as well as respiratory and neurological problems, compared with a similarly sized, unexposed control town in Mississippi (he declined to reveal the name of the town).

Dahlgren has published the only peer-reviewed studies on the health effects of people living near wood-treatment facilities. Two studies centered on towns in Mississippi—one in Grenada, which still operates a wood-treatment plant owned by Koppers Inc., and another in Columbus, home to a now-defunct facility owned by Kerr-McGee Corp. (which was purchased last year by Houston-based Anadarko Petroleum Corp.).

In each case, Dalhgren found that simply living near a wood-treatment facility caused elevated blood levels of several known carcinogens, including dioxins. General cancer rates were higher; upper respiratory problems and sinusitis were common.

Dahlgren expects cancer rates in Somerville will be far worse even than those in Columbus and Grenada, since the tie plant's operations were significantly larger and continued to use pesticides such as chromated copper arsenic and pentachlorophenol as recently as the late 1980s, long after the others had stopped because of widely reported health risks.

Dahlgren has already identified a dozen cases of stomach cancer in Somerville—which, he says, is as much as 40 to 60 times the national average for a town its size.

"Gastric cancer is rare for people in their 40s and 50s," says Dahlgren, who treated rescue workers exposed to toxins at the World Trade Center site on 9/11 and served as lead toxicologist in the famous Erin Brockovich case against Pacific Gas and Electric Co. "It's declining everywhere. But here, it's gone up."

Agency spokesman Clawson says the TCEQ "has no knowledge" of the recent environmental testing performed in Somerville.

The TCEQ has received 17 odor complaints against the tie plant since 2002 but has never conducted off-site testing to determine contamination levels for nearby residents. "We have not been able to determine the full extent of contamination," Clawson wrote in an e-mail. But he says his agency remains confident that most of the contamination is at the plant site itself, where it is being remediated.