Retraction Watch

Tracking retractions as a window into the scientific process

Authors retract study that found pollution near fracking sites

with 10 comments

Environmental Science and TechnologyThe authors of two environmental papers, including one about the effects of fracking on human health, have retracted them after discovering crucial mistakes.

One of the studies reported an increased level of air pollution near gas extraction sites, and the other suggested that 2010’s Deepwater Horizon oil spill in the Gulf of Mexico contributed to air contamination.

According to the corresponding author of both papers, Kim Anderson at Oregon State University, the journal plans to publish new versions of both papers in the next few days. In the case of the fracking paper, the conclusions have been reversed — the original paper stated pollution levels exceeded limits set by the U.S. Environmental Protection Agency (EPA) for lifetime cancer risk, but the corrected data set the risks below EPA levels.

The fracking paper received some media attention when it was released, as it tapped into long-standing concerns about the environmental dangers of hydraulic fracturing (fracking), which extracts natural gas from the earth. A press release that accompanied the paper quoted Anderson as warning:

Air pollution from fracking operations may pose an under-recognized health hazard to people living near them.

Both papers, published in Environmental Science and Technology, were retracted on the same day (June 29), both due to mistakes in reported levels of polycyclic aromatic hydrocarbons (PAHs), pollutants released from burning oil, gas, and other organic matter. 

Here’s the retraction notice for “Impact of Natural Gas Extraction on PAH Levels in Ambient Air:”

After publication the authors discovered a mistake in the air concentration calculations. PAH air concentrations reported in the original article are therefore incorrect. The calculation error resulted from using incorrect units of the ideal gas constant, and improper cell linkages in the spreadsheet used to adjust air concentrations for sampling temperature. Correcting this error changes air concentrations significantly relative to those reported in the published article. This correction also changes some of the conclusions reported in the original article.

Due to the impact of this correction on the reported findings, all authors retract the original article. The original article was published on March 26, 2015 and retracted on June 29, 2016.

The paper has been cited seven times, according to Thomson Reuters Web of Science.

The study was heavily criticized by Energy In Depth, a site launched by the Independent Petroleum Association of America, which argued:

…the study’s volunteers were spearheaded by a known anti-fracking group; the report completely ignores the likely higher-than-normal emissions from other sources; and, the researchers even admit that their work is “not statistically significant” because their sample size was too small and not random. Instead of garnering headlines promoting fear, the report should be notable for its glaring gaps in research.

According to Energy In Depth, the authors also ignored previous data that have suggested fracking is safe.

Regarding the retraction, Energy In Depth notes:

Ohioans deserve a full explanation as to why a study that generated numerous alarmist headlines by promoting fear was retracted. It will also be interesting to see if the retraction gets as much media attention as the flawed study generated.

(We’ve found one outlet — UPI — which has updated its story with news of the retraction.)

Anderson told us that a corrected version of the fracking paper is due to be published by the same journal soon, and sent us the following text from it:

Closest to active wells, the [excess lifetime cancer] risk estimated for maximum residential exposure was 0.04 in a million, which is below the U.S. EPA’s acceptable risk level. Overall, risk estimates decreased 30% when comparing results from samplers closest to active wells to those farthest. This work suggests that natural gas extraction is contributing PAHs to the air, below levels that would increase cancer risk.

Anderson told us that the original miscalculation of PAH levels was due to an “honest spreadsheet error,” adding:

It is important that we stand up and correct the record. This of course is hard as there are many as you are aware that will throw eggs and try to ruin careers, but it is the right thing to do. I very much regret the mistake, but felt it was important to set the record straight, no matter the cost.

Here’s the retraction notice for “Polycyclic Aromatic Hydrocarbon (PAH) and Oxygenated PAH (OPAH) Air–Water Exchange during the Deepwater Horizon Oil Spill:”

After publication the authors discovered a mistake in the air concentration calculations. PAH and OPAH air concentrations reported in the original article are therefore incorrect. The calculation error resulted from using incorrect units of the ideal gas constant and improper cell linkages in the spreadsheet used to adjust air concentrations for sampling temperature. Correcting this error changes air concentrations significantly relative to those reported in the published article. This correction also changes some of the conclusions reported in the original article.

Due to the impact of this correction on the reported findings, all authors retract the original article. The original article was published on November 20, 2014, and retracted on June 29, 2016.

The paper has been cited three times. 

If the journal does republish the two papers shortly, it resembles a trend we’ve seen in other journals by JAMA, which has recently begun to “retract and replace” articles affected by honest error, not misconduct.

The papers have four authors in common: Brian Smith, Lane Tidwell, Kevin Hobbie and Anderson, all based at Oregon State University.

Update: 7/11/16 10.43 a.m. eastern: An editor from the journal has informed us that the updated version of the Deepwater Horizon spill paper was published on July 8.

Update: 7/21/16 12.15 p.m. ET: The journal has now informed us that the updated version of the fracking paper was published on July 11.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

Comments
  • Timo Soren July 8, 2016 at 2:44 pm

    Let’s how the media coverage of this matches the origin PR.

    • Cheryl July 9, 2016 at 5:07 am

      Egg on All Gore’s face is richly deserved.
      I would also be interested in how this was reported in the media, and whether this new data was ever reported. I am so tired of hysterical media.

      • Marco July 9, 2016 at 11:01 am

        Also to you the question why there would be egg on Al Gore’s face?

  • james bancroft July 8, 2016 at 5:29 pm

    It seems, if it favors the destruction of western energy production and productivity, it is good to hear about, yet if it exonerates western energy production methods, then it might put egg on AlGore’s face, and therefore is bad.

    • Marco July 9, 2016 at 1:49 am

      Why would this “put egg on Al Gore’s face” ???

    • Tim Dibble July 12, 2016 at 12:46 pm

      Not everything is a conspiracy theory.

  • Steven McKinney July 8, 2016 at 8:53 pm

    “According to Energy In Depth, the authors also ignored previous data that have suggested fracking is safe.”

    Why, fracking is so safe that no one gets hurt by collapsing masonry in Oklahoma. Now that’s safe!

    Reuters:
    Money – Thu May 12, 2016

    Insurers shun risk as oil-linked quakes soar in Oklahoma

    “As the number of earthquakes in Oklahoma exploded into the hundreds in the last few years, nearly a dozen insurance companies moved to limit their exposure, often at the expense of homeowners, a Reuters examination has found.

    Nearly 3,000 pages of documents from the Oklahoma Insurance Commission reviewed by Reuters show that insurers and the reinsurers who cover them grew increasingly concerned about exposure to earthquake risks because of heightened frequency of seismic activity, which scientists link to disposal of saltwater that is a byproduct of oil and gas production.

    Even as they insured more and more properties against earthquakes in the past two years, six insurers hiked premiums by as much as 260 percent and three increased deductibles. Three companies stopped writing new earthquake insurance altogether, state regulatory filings obtained by Reuters show. Several insurers took more than one of those steps.

    In addition, the insurers would consider suing oil and gas companies for reimbursement in instances where they would have to pay damages to homeowners, according to several sources, including two insurance company officials.

    So far Oklahoma’s biggest earthquake was a 5.6 magnitude temblor in Prague in 2011 that buckled road pavement and damaged dozens of homes.

    However, the push to limit earthquake exposure reflects insurers’ fear that the surge in small quakes is a portent of a ‘big one’ in coming years, given the relationship between the magnitude and a total number of earthquakes in a certain area.”

    • Ben Vincent July 11, 2016 at 10:45 pm

      You are confusing fracking with the disposal of naturally brackish water was already underground naturally mixed in with oil and gas prior to any drilling. It was already underground before man did any drilling. This water is pumped out with the oil and gas and is then separated. It used to be left in man made surface evaporation ponds. This was deemed bad by environmentalists. So now oil companies return this water back underground. It is returned deeper than the wells for oil and gas extraction, for two reasons. By putting it back deeper there isn’t a risk of it contaminating much shallower fresh underground water. Also it doesn’t get pumped out, separated and pumped back underground again and again.
      Oil companies are working on ways to economically treat and purify this brackish water. If they can do that man will have a new source of fresh water.
      During fracking, oil companies recover and recycle and reuse as much of the fracking fluids as possible. It is not fracking fluid that is responsible for seismic activity.

      • Steven McKinney July 12, 2016 at 11:11 pm

        From the very same article

        “Scientists link the quakes to the injection of wastewater generated from the oil and gas production process deep underground. Volumes of so-called ‘produced water’ have ballooned as horizontal drilling and hydraulic fracturing, or fracking, boosted output in Oklahoma.”

        I’m not confused, you’re just nit-picking, and offering apologies to an industry that will go away, and the sooner the better.

        There’s plenty of sunshine in Oklahoma. Oklahomans need to get on with exploiting solar energy, instead of passing state laws inhibiting such activity, as their state crumbles around them.

  • Tim Dibble July 11, 2016 at 7:44 am

    The evaluation of safety is limited by the short term and statistical nature of the analyses. The imposition of fluids in the subsurface environment changes the relative (on a geologic scale) equilibria. Since the environment is subject to changes, but typically only on a geologic time scale, the fluids will create change, but the change is nearly impossible to predict on a human time scale. The content of the fluids may or may not affect humans in the human time scale, but the interactions and chemical reactions which occur in the subsurface under geologic conditions (think of the subsurface as a huge pressure cooker with all sorts of existing chemicals ready to react), still results in unknowable, immeasurable effects.

  • Post a comment

    Threaded commenting powered by interconnect/it code.