A landmark study by the U.S. Environmental Protection Agency that concluded fracking causes no widespread harm to drinking water is coming under fire—this time, from the agency’s own science advisers.

The EPA’s preliminary findings released in June were seen as a vindication of the method used to unlock oil and gas from dense underground rock. A repudiation of the results could reignite the debate over the need for more regulation.

Members of the EPA Science Advisory Board, which reviews major studies by the agency, said the main conclusion—that there’s no evidence that fracking has led to “widespread, systemic impacts on drinking water”—requires clarification, according to an email from David Dzombak, a Carnegie Mellon University environmental engineering professor leading the review. The panel Dzombak heads will release its initial recommendations later this month.

“Major findings are ambiguous or are inconsistent with the observations/data presented in the body of the report,” the 31 scientists on the panel said in December in a response to the study.

The scientific panel’s recommendations aren’t binding, and the EPA is not required to change its findings to accommodate them. But they already are raising questions about the most comprehensive assessment yet of a practice that has driven a domestic oil and gas boom but also spawned complaints about water contamination.

An EPA spokeswoman said the agency will use comments from the scientists and the public to “evaluate” possible changes to the report.

Oil Exports

A significant change could be a big blow to an industry that is celebrating major policy wins, including the end of trade restrictions that for four decades blocked the export of most raw, unprocessed U.S. crude.

Fracking, also known as hydraulic fracturing, involves pumping water, sand and chemicals underground to free oil and gas trapped inside dense rock formations.

For the study, mandated by Congress, the EPA analyzed more than 950 sources of information, including previously published papers, state reports and the agency’s own scientific research, but found no clear evidence that the fracking process itself could cause chemicals to flow through underground fissures and contaminate drinking water. When the agency took a broader look at the entire water cycle around fracking—from getting water supplies to disposing of fluid waste—it documented instances where failed wells and above-ground spills may have affected drinking water resources.

Peer Review

Robust peer review by the EPA’s Science Advisory Board, established by Congress in 1978, is designed to ensure the integrity of scientific reports, agency spokeswoman Melissa Harrison said in an email.

She said the agency will use the comments from the advisory panel, as well as those submitted by the public, “to evaluate how to augment and revise the draft assessment.”

“The final assessment will also reflect relevant literature published since the release of the draft assessment,” she said.

Texas, Wyoming

Environmentalists want the final document to include more information about alleged contamination near drilling sites in Dimock, Pa.; Parker County, Texas; and Pavillion, Wyo. Those episodes “show how out of step the conclusion is with the body of the report,” Clean Water Action oil and gas campaigner John Noel said in an interview.

The scientific review panel seems intent on suggesting changes—at one point floating the idea of “explicit” descriptions of what happened in Dimock, Parker County and Pavillion. The group could ask the EPA to rescind its top-line finding altogether or clarify it by asserting that the lack of widespread, systemic impacts from fracking is relative to the number of wells drilled.

When one of the panelists—University of California Engineering Professor Thomas Young—suggested such a rewrite during three days of meetings last October, the group broke out in spontaneous applause.

‘Science Settled’

Industry lobbyists and trade groups are working to tamp down the panel’s criticism, with American Petroleum Institute President Jack Gerard casting it as the work of determined environmental activists opposed to fossil fuels.

“The science should be settled,” Gerard told reporters at a news conference last week. “There are a handful of people who are not happy with the outcome, and they continue to drive their agenda based on ideology, not based on the science.”

The API and the Independent Petroleum Association of America delivered a similar message in separate letters to the EPA.

Scott Segal, head of the policy resolution group at Bracewell & Giuliani in Washington and a lobbyist who represents Range Resources Corp. and other energy companies, said in an interview that the review board should disregard “anecdotal evidence presented by litigants in active cases.” By contrast, he added, “the overwhelming weight of scientific evidence is on the side of the regulated community.

Several of the EPA’s science advisers reviewing the fracking report said the evidence doesn’t support the conclusion about water safety.

Spill Data

Spill data alone “gives sufficient pause to reconsider the statement” that there’s no evidence of systemic, widespread damage, said panelist Bruce Honeyman, professor emeritus at the Colorado School of Mines.

“It’s important to characterize and discuss the frequency and severity of outliers that have occurred,” said panelist Katherine Bennett Ensor, chairwoman of the Rice University Department of Statistics.

And panel member James Bruckner, a professor of pharmacology and toxicology at the University of Georgia, said the report glosses over the limited data and studies available to the agency.

“I do not think that the document’s authors have gone far enough to emphasize how preliminary these key conclusions are and how limited the factual bases are for their judgments,” Bruckner said.

Young, the University of California professor who suggested rewriting the top-line conclusion, faulted the document for trying “to draw a global and permanent conclusion about the safety or impacts of hydraulic fracturing at the national level” given the “uncertainties and data limitations described in the report.”