The NMHCs in the study were detected at levels of parts per million, parts per billion and parts per trillion, but the endocrine system is so sensitive that even tiny doses can lead to large health effects. Federal safety standards rarely consider the impacts of low dosage testing, an omission that scientists say should be addressed.
The study's authors detected thirty NMHCs that affect the endocrine system. Several belong to a class of compounds called polycyclic aromatic hydrocarbons (PAHs) and were detected at levels that other scientists have found are high enough to impact child development. In those studies, clinical researchers gave pregnant women living in cities personal air monitors, then tracked their children's development. Women exposed to a certain level of PAHs were more likely to have children with lower birth weight and lower IQ scores.
One chemical found at surprisingly high levels was methylene chloride, a common laboratory solvent. It's not a component of raw gas and doesn't appear on any of the public disclosure forms of chemicals used during drilling and fracking. "However, residents and gas field workers have reported that methylene chloride is stored on well pads for cleaning purposes," the authors wrote.
Robert Howarth, a Cornell University scientist who wasn't involved in the study, said the presence of methylene chloride points to a need for better chemical disclosure laws. "Methylene chloride is a surprise…We need a lot more information on what's used at drilling sites overall."
While drilling companies are required to disclose many of the chemicals used for fracking, they are usually allowed to keep proprietary chemicals secret. Drilling and cleaning compounds are rarely, if ever, subject to public disclosure.
Sgamma, the industry representative, said she was not aware of methylene chloride being used on well pads. She said the samples were probably contaminated in the lab.
Kwiatkowski said TEDX considered that possibility and ran blank samples to test for contamination. They didn't find methylene chloride in the blanks, but found it "over and over again" in the collected air samples.
The TEDX study was inspired by years of complaints about headaches and respiratory problems the researchers had heard from people living near gas wells. Many of the symptoms began the moment drilling started, long before the wells were fracked, Kwiatkowski said.
That prompted the scientists to study air quality before and after drilling. They identified a well pad slated for drilling and set up an air sampling station near a home 0.7 miles from the well pad. There were 130 active gas wells within one mile of the site.
Kwiatkowski said the ideal sampling station would have been located directly on the well pad, but TEDX has had little success persuading the industry to cooperate with its research, so the researchers didn't ask for access to the well pad. They also didn't want to draw unwanted attention to their work. Local residents are divided when it comes to the benefits and risks of gas drilling, and Kwiatkowski said they didn't want to cause trouble among neighbors. Their choice of location was further constrained by the need for a constant source of electricity and the need to protect the station from possible vandalism.
The scientists took a partial set of baseline data on July 2010, before any wells had been drilled on the pad. On October 19, after residents called to report activity on the well pad, the scientists rushed in to take a full set of baseline readings. The first well was drilled three days later.
Air sampling continued weekly until October 2011. All samples were analyzed in EPA-certified labs. The scientists tested for more than 100 chemicals and found over 50 at levels high enough to be detected by their instruments.
When the dates of the drilling and fracking activity were posted online, in accordance with Colorado's disclosure laws, the scientists learned that the company had drilled 16 wells on the well pad between October 2010 and March 2011. Two other well pads were drilled starting in April and July. Fracking followed the drilling. About 100 other wells within a mile already were producing gas and were neither fracked nor drilled during the study period.
The data showed a major spike in chemical concentrations after the first 16 wells were drilled, but not after fracking. The increase was significant when compared with the baseline samples collected before the drilling, as well as samples from most of the year after drilling stopped.
Colborn said that suggests the increased emissions are linked to the raw gas released from drilling—but she said there's no way to tell for sure, because they couldn't directly sample emissions from the well pad. Colborn said TEDX and other scientists are already making plans for a follow-up study to chemically fingerprint the source of the pollutants.
Jackson, the Duke University scientist, said the paper hasn't convinced him that the increased emissions are directly tied to the well pad it studied instead of the combined effect of the region's natural gas operations.
He said the evidence is weak because the spikes occurred only during the middle two months of the five-month drilling period, and because the emissions could have originated from the 130 other wells in the region.
There's no question the study "is documenting air quality in that valley," he said, "and that's still valuable," especially when it comes to health implications for local residents.
Industry Questions Study's Credibility