The Environmental Protection Agency’s landmark investigation into the impact of fracking on drinking water lacked baseline testing that would have made its results more illuminating, according to a scientific panel that assessed it, and independent analysts.
The Hydraulic Fracturing Research Advisory Panel, a unit of the EPA’s Science Advisory Board (SAB), published its evaluation of the EPA’s report on Jan. 7.
The panel said the EPA’s report excluded “prospective case studies”, also known as baseline testing, in which water quality is assessed before drilling takes place to determine whether subsequent natural gas development has an impact on water sources.
“Such studies would allow the EPA to monitor water acquisition and its effects to a level of detail not practiced by industry or required by state regulation,” the SAB report said. “Such detailed new data would allow the EPA to reduce current uncertainties and research gaps about the relation between hydraulic fracturing water acquisition and drinking water.”
Dave Yoxtheimer, an extension associate with Penn State University’s Marcellus Center for Outreach and Research, agreed that the inclusion of baseline cases would have aided the assessment of whether fracking contaminates drinking water but that didn’t happen, reportedly because many drillers would not allow EPA inspectors to take baseline samples.
“Even a generalized analysis of exposure scenarios using common technologies and common potential contaminants would be a helpful exercise,” Yoxtheimer wrote in an email. “If the prospective studies were conducted as originally proposed then this would have afforded an opportunity for exposure studies to have been conducted at those test sites.”
Although baseline testing is not technically required in Pennsylvania, the state’s oil & gas law, Act 13, assumes that a company that does not sample water before drilling is liable for any contamination of a water well within 2,500 feet of a shale well or if contamination occurs within 12 months of a well being drilled, Yoxtheimer said.
As a result, most energy companies operating in Pennsylvania now routinely perform baseline testing, he said. In other states, the practice is becoming more common, although it is not yet universal.
Seneca Resources, which is active in the Marcellus Shale of northwestern Pennsylvania, supports baseline testing because it can reveal that private water wells are contaminated with a range of pollutants, especially sewage, not just fracking-related material, said Rob Boulware, a spokesman for the company.
“It’s important for all landowners to have their wells tested,” Boulware told StateImpact. “Many things could affect water.”
Nadia Steinzor, Eastern Program Coordinator for Earthworks Oil & Gas Accountability Project, said a lack of baseline tests can make it easier for operators to deny that fracking has contaminated water.
“The lack of baseline data is often something that both operators and regulators hide behind when they are reaching conclusions about impact to water,” Steinzor said.
If such testing does take place, its effectiveness can be reduced by a disconnect between energy companies and regulators over baseline standards, she said.
“You end up with an apples-and-oranges situation which often gives operators cover to say that there were no impacts,” she said. “The SAB is completely correct to say that this was one of the greater limitations in the EPA report.”
The comments on baseline testing were part of a report that concluded there was no evidence of widespread or systemic water contamination from fracking, a conclusion that was welcomed by industry and attacked by critics, who accused the EPA of downplaying evidence of contamination.
Yoxtheimer said the EPA’s statement that it had found no evidence of widespread impact on water supplies may have been misinterpreted as a “green light” to proceed with fracking.
But that conclusion would ignore the fact that local impacts have been, as the SAB pointed out, severe, and so a better interpretation of the EPA’s report would have been to proceed but with caution, Yoxtheimer said.
“I believe the point the EPA is trying to make is that not every well is problematic, that in fact a small percentage of wells represent a risk to water,” he said. “However, it is potentially a big issue on a local scale when something goes wrong in the process.”
He said the EPA’s conclusion about no widespread impacts “glosses over” the problems that have occurred on a local basis.
The energy industry has also criticized the lack of baseline testing.
The American Petroleum Institute hired Battelle Memorial Institute, a nonprofit science and technology research organization, to evaluate the EPA’s plan at an early stage.
In 2011, Battelle criticized the plan on a number of grounds, including the argument that the lack of baseline testing would undermine the study’s scientific credibility.
“The limited and possibly statistically biased pool from which the retrospective sites were drawn, and the lack of baseline information, are likely to limit the scientific validity and usefulness of case study findings and may result in incorrect or flawed conclusions,” Battelle said.
Meanwhile, Briana Mordick, a blogger with the Natural Resources Defense Council, said the EPA’s statement about no widespread impacts, contained in its executive summary, misrepresents the science contained in the body of the report, and should be revised.
“The EPA’s conclusion seriously misrepresents the findings of the study and the uncertainties and major data gaps that the EPA found again and again,” Mordick wrote in a blog responding to the SAB’s report. “We agree with the SAB Advisory Panel that EPA should revise it in order to accurately communicate the science.”
For its part, the SAB said the EPA needs to provide a lot more detail about the impacts on locations like Dimock in Susquehanna County, where some residents have blamed the gas industry for water contamination.
The SAB report also accused the EPA of providing “limited information” about the frequency and nature of spills of fracking fluid at well sites.
The panel agreed with EPA that a national extrapolation on spill risks can’t be based on observations from two states, and that there are important gaps in publicly available data on the withdrawal of water for fracking.
But it called for more information on the intensity and duration of human exposures to fracking activities to determine whether the industry is affecting health in specific locations.
Julia Valentine, an EPA spokeswoman, declined to elaborate on the agency’s initial reaction to the SAB report, which said the fracking report would be revised in light of comments from the SAB and the public.