Earlier this month, a Bloomberg News headline made a very serious accusation about hydraulic fracturing and shale development: “Study Shows Fracking Is Bad For Babies” was the title of an opinion column by Mark Whitehouse. The opening paragraph followed with a strong indictment of the oil and gas industry:
“The energy industry has long insisted that hydraulic fracking — the practice of fracturing rock to extract gas and oil deep beneath the earth’s surface — is safe for people who live nearby. New research suggests this is not true for some of the most vulnerable humans: newborn infants.”
There is an odd parsing of words here – the headline says the study “shows” the process “is” bad for babies, while the lead paragraph went with the more nuanced “research suggests.” But the op-ed was clearly attempting to highlight that shale development not only could harm public health, but indeed already “is.”
Yet it’s not until the third paragraph before Whitehouse admits that the study “has yet to be peer-reviewed or posted online.” Bloomberg News hosted the publication of preliminary findings, and gave the piece a dubious headline that indicates what’s reported in the op-ed is an established fact.
After the op-ed was published, however, the study’s authors spoke up and undermined the premise of the headline, if not the entire column. One of the authors, Michael Greenstone of MIT, told the New York Times’ Andy Revkin that coverage of the study may have jumped the gun:
“The newspaper articles describe preliminary results that we did not intend to share with press. We will release a full working paper as soon as we are finished with the analysis.” (emphasis added)
Janet Currie of Princeton University, another author, provided even more pushback:
Another study author, Janet Currie of Princeton, said this study is not even a “working paper” yet: “[W]e are not trying to publicize the paper ahead of peer review. We have not put the paper out as a working paper and aren’t comfortable circulating it yet.” (emphasis added)
Those statements were made within days of the op-ed’s publication, and yet Bloomberg has refused to pull the op-ed or even adjust the headline for accuracy.
This is indicative of a broader problem, and it’s certainly not limited to Bloomberg News. Claims that suggest harm from hydraulic fracturing and shale development are all too often not receiving the scrutiny they deserve. Allegations against shale development are automatically assumed to be credible, while regulatory determinations and scientific studies affirming safety are dismissed or glossed over.
Even worse is the fact that some of these critiques are printed in the context of scientific analysis, when the reality is much more nuanced and far less conclusive than what is conveyed.
Parker County, Revisited
In addition to Bloomberg’s willingness to print “preliminary results” as a definitive indictment of shale’s impact on children, others have fallen victim to the same urge to hype accusations before proper review.
A recent story from the Associated Press suggested that “new tests” link explosive methane concentrations in groundwater to gas drilling activity in Parker County, Texas. The case itself has been extensively covered, and has become something of a cause célèbre for anti-fracking groups across the country. Activists infamously worked with the EPA regional administrator back in 2010, Al Armendariz, to “get jurisdiction to EPA” in the case, and Armendariz obliged them with a costly endangerment order against the operator, Range Resources. Geochemical testing from Weatherford Labs ultimately revealed that the methane was naturally occurring, and Range was exonerated by state regulators based on that evidence. (The EPA failed to test for nitrogen, the presence of which distinguishes naturally occurring methane in the Strawn formation from Barnett gas.) EPA was forced to drop the order in 2012, but not before Range suffered more than a year of reputational harm over a baseless enforcement action.
The recent AP story, however, ignored the data that proved Range was not responsible and focused on a “preliminary analysis” from a researcher who claims the gas was from Range’s operations. The AP also relied on a new analysis from Duke University’s Rob Jackson, albeit with a caveat that did not deserve to be buried, as it was:
“Jackson, the Duke University professor, also specializes in isotopic analysis. He declined to share his study — funded by Duke and the National Science Foundation — until it is peer-reviewed and published, but some homeowners shared test results with the AP.” (emphasis added)
Let’s ignore the blatant funding omission (Jackson’s work is funded by Duke’s Nicholas School of the Environment, which has received $120,000 since 2011 from the anti-fracking Park Foundation), as well as many of the same oversights that undermined the reporter’s previous coverage of the topic. We still have yet another example of the media latching on to unpublished, non-peer-reviewed, and preliminary analyses to push a story about the “risks” of development.
The researcher himself (Jackson) refused to share the results because they had not yet gone through appropriate scientific scrutiny, and yet the AP reporter felt comfortable sharing them with the world in a national wire story. Oddly, although Jackson “declined to share his study,” he still spoke to the AP at length about the findings. If he’s willing to talk about his study in detail, did he really refuse to release it? Are we only getting one side of a more nuanced picture, as was the case with the “bad for babies” study?
Cornell Caught in the Act
A similar phenomenon played out in the summer of 2012, when a Cornell graduate student released a study suggesting a correlation between hydraulic fracturing and low birth weight. The results were blasted out to the media, and several news outlets cheerfully used the study to write another story about the risks of development. Activists from across the country piled on, using the research to make inflammatory claims about how hydraulic fracturing harms children.
But a deep dive into the research, cataloged on the New York Times’ Dot Earth blog, revealed that a PR firm had pitched the study to reporters – without disclosing that the firm had been hired by New Yorkers Against Fracking to promote the findings.
The author of the study, Elaine Hill, was also asked about the study and whether it had been peer-reviewed. It had not gone through that process, and Hill herself was forced to walk back her definitive statements to the press. She originally said her research was “robust” and indicated that “future generations may be seriously harmed” from hydraulic fracturing. But after receiving criticism from the scientific community for publicizing her results at an anti-fracking forum before they had been properly reviewed, Hill backtracked by saying: “I do not want to be in the policy spotlight so it is in my best interest to back out of the conversation now and focus on my work.”
The Times‘ Revkin suggested this is what can happen “when publicity precedes peer review.”
One of the more interesting critiques was from Lisa McKenzie of the Colorado School of Public Health, whose observation has relevance far beyond Hill’s work specifically: “Any results released prior to peer review,” McKenzie said, “should be approached with extreme caution.”
Hysteria v. Science
There are many other examples of claims and headlines diverging from the scientific conclusions they are attempting to categorize. This includes the inevitable headline linkage between “fracking” and earthquakes every time a researcher suggests wastewater disposal (a separate process from fracking) may be connected to seismicity (those studies typically are peer-reviewed, but the misrepresentation of results to malign “fracking” is certainly in the same league as the preceding examples). Many health impacts are often improperly tied to hydraulic fracturing as well — although, to its credit, the AP recently exposed some of these as examples of bad science.
What ties all of this together is an increased perception of risk about a process that is highly technical and often difficult to understand. Indeed, recent polling suggests hydraulic fracturing is not well understood by most of the general public. As such, any claims made about the process – either supportive or critical – can shape opinions much more easily than claims made about more familiar subjects.
This raises several important questions: What problem are we solving by publishing inflammatory claims about shale development when those claims have not been properly vetted? Are we attempting to understand the complexities of development, or simply pushing stories that will attract attention? Are we focusing on risk management, or simply scaring people?
U.S. shale development has transformed our domestic energy equation, and there is plenty of evidence that it will transform the global energy picture, if it has not done so already. States like North Dakota and Texas weathered the recent recession better than most due to increased oil and gas development from shale. Billions of dollars are flowing to farmers and other landowners who have struggled financially for decades. Hydraulic fracturing and horizontal drilling have helped unlock an enormous supply of natural gas, putting downward pressure on consumer prices and saving households billions of dollars in energy costs. Increased natural gas use has also helped the United States cut its carbon dioxide emissions to a 20-year low, while increased domestic oil production has reduced imports to their lowest level since the early 1990s.
But there are risks that accompany shale development, and there is plenty of room to debate how best we can manage those risks. Additional research can help us better understand both risks and benefits, so we should all welcome new scientific data that inform us of ways to further improve safety. With the benefits of shale development as significant as they are, we should all advocate for a responsible path forward that also allows those benefits to accrue.
Unfortunately, the rush to associate “fracking” and shale development with a variety of harms — before the claims have been properly vetted or reviewed, and often without nuance — is completely inconsistent with that goal. The fact that many of those claims have proven later to be baseless or exaggerated demonstrates precisely why accuracy should always trump allegation.