Hazard of Extraction
America’s ‘fracking boom’ is boosting local economies and lessening our dependence on on fuel imports, but at what price?
The Colorado River begins its 1,450-mile journey high on the continental divide. It tumbles southwest through Glenwood Canyon, where cathedrals of rock carved by the river soar above Interstate 70. The highway, which runs through a spectacular 12.5-mile gorge, is an engineering marvel.
forty miles downstream, the river and I-70 run side-by-side through another remarkable feat of engineering. Some 10,000 drilling rigs dot the landscape of Colorado’s Grand Valley, tapping into vast reserves of natural gas. The reserves, which lie deep within the region’s Piceance Basin, are now accessible and profitable to tap thanks to advances in hydraulic fracturing.
But while hydraulic fracturing has ushered in an oil and gas boom, some scientists are questioning its effects on public health. Susan Nagel, associate professor of obstetrics, gynecology and women’s health at the MU School of Medicine, is one of them. She’s the lead author of “Estrogen and Androgen Receptor Activities of Hydraulic Fracturing Chemicals and Surface and Ground Water in a Drilling-Dense Region,” which recently appeared in the journal Endocrinology. The study found that water near hydraulic fracturing sites had greater levels of endocrine-disrupting chemicals (EDCs) than in areas without drilling. Exposure to these chemicals, even in very low doses, has been linked to cancer, birth defects and infertility.
The study generated a flurry of attention when it was first published online last December. Along with news outlets in Missouri and Colorado, publications such as National Geographic and The Los Angeles Times carried stories about its implications.
“We certainly know that exposure to endocrine-disrupting chemicals during fetal life and early life can alter behavior in kids — such as activity and rough-and-tumble play,” says Nagel. “If your baby is developing in this environment, you only get one shot at that. You can add a whole lot of things to a cake while you’re mixing it up, but once it’s out of the oven, it’s cake.”
Nagel, who also holds a position in MU’s Division of Biological Science, has spent her 20-year career studying how EDCs can disrupt the delicate balance of the endocrine system. This system regulates a number of body functions, including growth, response to stress, sexual development and metabolism. Nagel has been particularly interested in fetal programming: the idea that things like chemical exposure or maternal health and diet can have a lifelong effect on offspring.
With that backdrop, and with the rapid rise in natural gas extraction, Nagel began tuning into concerns associated with hydraulic fracturing. “In my mind, it was a potential source of human and wildlife exposure to endocrine disruptors,” she says.
But researching the health effects of hydraulic fracturing can be difficult, because many of the chemicals used in the process are proprietary. Hydraulic fracturing, also known as fracking, involves blasting millions of gallons of water mixed with a slurry of chemicals and lubricants deep underground to crack shale formations and release oil and gas. The process is exempt from some parts of the Safe Drinking Water Act, and companies don’t have to disclose the chemicals if they’re considered a trade secret. All in all, there are about 750 chemicals known to be used in fracking.
“In 2011, Congress asked eight, then 14 oil and gas companies to voluntarily disclose the chemicals used in hydraulic fracturing,” says Nagel. “The 750 chemicals were a tabulation of the lists that those oil and gas companies turned in to Congress. It doesn’t mean that all companies are necessarily using all those chemicals.”
To get more information on the chemicals and how they might be tested in the lab, Nagel turned to The Endocrine Disruption Exchange (TEDX), a group of scientists and researchers who share data and analyses of studies on EDCs. “They had already done a lot of investigative work to determine what chemicals were used in Colorado,” she says. “Certainly for some of the chemicals there’s actually a lot of information, and exposure is already known to be associated with risks to human health.”
Researchers and scientists associated with TEDX also had anecdotal information from residents in Colorado who lived near wells where there had been spills. For spills of a certain size, oil and gas companies are obligated by law to report, fix and then test spill sites for the presence of a few select chemicals that are a hallmark of contamination. But some residents reporting changes to ground and surface water didn’t feel that their concerns had been addressed.
“The public reported seeing surface conditions — bubbling and fizzing in a creek.” says Lisa Bracken, whose property in Silt, Colo., was affected by spills in 2004 and 2008. Bracken says that the creek could be lit on fire and believes that her father developed pancreatic cancer after drinking from it. “He drank from West Divide Creek when it was flowing with benzene (unbeknownst to us at the time),” she writes in an email. The creek is a tributary to the Colorado River.
“I believe fracking chemicals should be disclosed and scrutinized,” says Bracken. “They should not be protected — especially if they enter the public water supply. The river and the wildlife speak for themselves, but in voices that few can hear or understand.” She has created a website to voice additional observations and concerns.
But Nagel says it’s one thing to say there’s a problem and quite another to look at a potential mechanism of action and test it in the lab. That’s where her research came in. Using chemical information from TEDX, she and her team designed a two-part study.
First, they tested 12 suspected or known EDCs, measuring their ability to mimic or block the effects of estrogen and androgen. The results showed that all 12 of the chemicals had some form of hormone-disrupting activity. Eleven blocked estrogen, 10 blocked androgen and one mimicked estrogen.
“We chose chemicals that were used in Colorado and that were used the most frequently,” she says. “That is, as self-reported by the oil and gas companies.” They also had to be commercially available.
In the study’s second part, Nagel’s team collected 39 unique samples of ground and surface water. Ground water is generally considered well water, while surface water sustains wildlife and ecosystems. Collection sites included: five sites in the Grand Valley, Garfield County, Colo., where hydraulic fracturing spills had been documented; two sites in Garfield County where there had been little or no drilling; nearby portions of the Colorado River; and three sites in Boone County: the Missouri River, Hinkson Creek and the USGS’s Columbia Environmental Research Center.
Results indicated moderate-to-high levels of endocrine-disrupting activity in the five drilling-spill sites, and moderate levels in the samples taken from the Colorado River. By contrast, areas in Garfield County, where there was little drilling, and in Boone County, where there was no drilling, showed low levels of EDCs.
Along with finding EDCs in surface water, the reserachers also found some differences in ground water, which wasn’t predicted.
“We tested for the ability of water samples to disrupt sex hormones in the lab, using human cells,” says Nagel. “We did not test the water for the presence of those 12 chemicals. That is a limitation, and we know the limitations of our study better than anyone.”
The approach does have two advantages, she says: It doesn’t rely on knowing the identity of all the chemicals in fracking fluids, and it’s a good way to measure the chemicals’cumulative effects.
“That approach has absolutely been used before. It has not been used before in this context, and it’s not the classical way that the oil and gas industry would look for water contamination — which generally involves testing for a limited number of select chemicals only.”
If Nagel sounds defensive, it might be because groups with ties to the oil and gas industry have been strident in their criticism of her study. The industry blasted her, for example, for saying in a press release that fracking is exempt from federal regulations. “So to be precise, it is exempted from parts of six of eight key federal regulations,” she says.
In addition, a post on the Energyindepth website, sponsored by the Independent Petroleum Association of America, called the study “inflammatory.” The post listed five problems, one of which involved collecting samples at spill sites.
“They said, ‘We know spills are bad, so what is this really telling us?’But not very long ago, and still very much today, they claimed that spills didn’t get into the ground water,” says Nagel.
Spills associated with hydraulic fracturing are not rare. According to the Colorado Oil and Gas Conservation Commission, more than 500 were reported last year in Colorado alone. In an effort to allay public concerns, Anadarko Petroleum Corporation and Noble Energy created the non-profit Coloradans for Responsible Energy Development. CRED has spent heavily on television ads in Colorado touting the safety of hydraulic fracturing.
Many in the public health community consider Nagel’s study a stepping stone, one that will provide an incentive to supplement today’s data with more findings. Laura Vandenberg, assistant professor of public health at the University of Massachusetts-Amherst, points out that the study did not collect water samples both before and directly after the fracking activities began, so it’s not possible to conclude that hydraulic fracturing was the cause of the water’s hormonal properties. However, she says that the authors were careful in how they worded their conclusions, and she doesn’t think they over-reached in their discussion of the implications of their results.
“This to me highlights that a new study should be done,” writes Vandenberg in an email. “New fracking sites are being developed, and water samples should be collected from surface and ground sources prior to any drilling activities. Overall, I think this study highlights another issue that the public health community needs to consider in the larger discussion about fracking. There are many reasons why natural gas is preferred to other sources of energy like coal, but we cannot ignore these unexpected and unintended consequences of fracking.”
Carol Kwiatkowski, assistant professor adjunct of integrative physiology at the University of Colorado, is also thinking about the broader merits of the research. “This is a very important study because people are just beginning to look at endocrine-related health effects from chemicals in natural gas operations,” she writes in an email. “No single study can answer all the questions we would like to ask. This is an important first step in determining the hormonal activity of water sources near natural gas operations.”
However, Larry Wolk, executive director and chief medical officer of the Colorado Department of Public Health and Environment (CDPHE), did not call for further inquiry. His department completed a brief analysis and issued the following statement in an email.
“There are numerous (thousands) septic systems in Garfield County. We don’t know how this may influence endocrine-disrupting chemical concentrations in groundwater.
“We question whether comparing EDCs in groundwater from Missouri with Garfield County, Colorado is legitimate. The study lacks concentration data, so it is difficult to respond to what the study means compared to established water quality standards. Some of the EDCs in the study have standards.
“Some sample sites are associated with spills. Therefore, the sample results should reflect a fairly localized area and not the entire unconfined aquifer. And, there is no indication in the study that any of the sample sites are currently used for drinking water.
“Comparing gas and fluid migration from the Marcellus Shale to Garfield County is not factually or scientifically valid.”
Leslie Robinson, chair of the Grand Valley Citizens Alliance, responded to Wolk’s statement in an email, citing oil and gas politics. The alliance does not trust the CDPHE to “ protect our health in the Gas Lands, because they seem to be more concerned with defending the health of the O& G industry,” she writes.
For her part, Nagel says that her team did not compare fluid migration from the Marcellus Shale to Garfield County; they cited research that demonstrates there is a deep connectivity between the shale gas layer and shallow ground water in the Marcellus shale. She adds that the study did not compare groundwater from Missouri with Garfield County; they compared ground water at spill sites and non-spill sites in Garfield County.
Finally, John Adgate, a professor and department chair of environmental and occupational health at the Colorado School of Public Health, takes a neutral approach. He says that, although he is not familiar enough with the study to comment on it specifically, more research in this area is needed.
Nagel has been somewhat taken aback by the attention her work has received. “It is a bit surprising that a small study is so noteworthy,” she says. “I guess I tend to think it’s a really good example of how little we know about this.”