Researchers at the University of Texas at Arlington’s (UTA) Collaborative Laboratories for Environmental Analysis and Remediation (CLEAR) recently released three new reports examining the diversity of bacteria found in water wells in the Eagle Ford and Barnett Shales. These reports claim that potential contamination from drilling activities impacted the bacteria found in the groundwater – bacteria that could gravely impact people’s health.

However, despite the various media reports – and even UTA’s own press release – suggesting the study showed “dangerous bacteria” thriving near oil and gas development, the study clearly states that these findings show no clear connection between fracking and the bacteria. As one study, titled Exploring the links between groundwater quality and bacterial communities near oil and gas extraction activities, states:

“Very little evidence of groundwater contamination was found that could be attributed to conventional and/or unconventional oil and gas development, with the exception of FC11 and three alcohols that were detected. While the sampling sites were all surrounded by UD activities, UD production wells do not contribute to groundwater contamination in a systematic fashion (Hildebrand et al., 2017, 2015). (emphasis added)

The study continues:

“As such, these data do not provide a definitive link between UD or agricultural activities and the groundwater microbiome; however, they do provide a baseline measurement of bacterial diversity and quantity in groundwater located near these anthropogenic activities.” (emphasis added)

Even one of the study’s authors, Kevin Schug, directly admitted that this study does not provide a link between fracking and contaminated water. As the Dallas Morning News reports:

“Shug, one of the authors and director of UTA’s Collaborative Laboratories for Environmental Analysis and Remediation, or CLEAR, said the research doesn’t conclusively connect fracking with the contaminated water. Naturally occurring natural gas, agriculture waste or some combination could be the contamination sources.” (emphasis added)

Sampling Issues

With that said, there are some methodological concerns with these studies that should be addressed; namely, the lack of control samples and extremely small sample sizes.

One of the studies – Exploring the links between groundwater quality and bacterial communities near oil and gas extraction activities (referred to as “Study 1” in this blog) – focused on water wells in the Eagle Ford, and relied on samples from only 19 water wells in a single county, while the two other studies analyzed just seven water samples from a previous study. None of the studies include control samples to examine how the findings compare to water not taken from areas near drilling activity.

Further, the seven samples used in two of the studies – Characterization of bacterial diversity in contaminated groundwater using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (“Study 2”) and MALDI-TOF MS for Identification of Cultivatable Organic-Degrading Bacteria in Contaminated Groundwater near Unconventional Natural Gas Extraction Sites (“Study 3”) – were chosen from a previous sampling taken for a 2016 study by UTA specifically because they were “abnormal.” As Study 3 states:

“The presented samples were chosen due to their abnormal levels of organic pollutants, as shown in the Results section of this manuscript (Section 3.1).”

Compounding this sampling issue, none of the samples were studied blind. In fact, two of the samples were taken “from private water wells characterized by their respective owners as anecdotal claims of UD-related contamination.”

Curious Framing

As shown, the authors admit that these studies do not show a definitive link between fracking and water contamination. And yet, oil and gas development is constantly pointed to as the originator of the water contamination, with the authors seemly downplaying other possible factors of water contamination such as agricultural activities. This is especially confusing considering Study 1 acknowledges that farming is common in the rural area sampled, but only “oil and gas extraction activities” are featured prominently throughout – including the headline. As the Study 1 abstract reads:

“In this work, groundwater samples were collected from a rural area in southern Texas, where agricultural activities and unconventional oil and gas development are the most prevalent anthropogenic activities.”

But even with this acknowledgement, Study 1 proceeds to list how “several” environmental studies have suggested that fracking is a cause of contamination – without doing the same for the agricultural activities mentioned:

“Several environmental studies have postulated that unconventional oil and gas development processes (UD), including hydraulic fracturing, may change the chemical composition of groundwater overlying hydrocarbon-rich petroliferous strata (Fontenot et al., 2013; Hildenbrand et al., 2017, 2016, 2015) and may affect the microbial communities that they support. Previous investigations have examined the impacts of hydraulic fracturing on surrounding environmental microbiomes in headwater stream ecosystems and surface waters (Fahrenfeld et al., 2016; Trexler et al., 2014). In both studies, the authors revealed that the microbial communities changed in response to altered conditions due to UD activities.” (emphasis added)

More concerning still, the other two studies take water contamination from oil and gas activities as almost an inherent truth, with the authors including oil and gas development not once, but twice as the potential cause of contamination at the very beginning of Study 2’s abstract:

“Groundwater is a major source for drinking water in the United States, and therefore, its quality and quantity is of extreme importance. One major concern that has emerged is the possible contamination of groundwater due to the unconventional oil and gas extraction activities. As such, the impacts of exogenous contaminants on microbial ecology is an area to be explored to understand what are the chemical and physical conditions that allow the proliferation of pathogenic bacteria and to find alternatives for water treatment by identifying organic-degrading bacteria. In this work, we assess the interplay between groundwater quality and the microbiome in contaminated groundwaters rich in hydrocarbon gases, volatile organic and inorganic compounds, and various metals. Opportunistic pathogenic bacteria, such as Aeromonas hydrophila, Bacillus cereus, Pseudomonas aeruginosa, and Stenotrophomonas maltophilia, were identified, increasing the risk for consumption of and exposure to these contaminated groundwaters. Additionally, antimicrobial tests revealed that many of the identified bacteria were resistant to different antibiotics. The MALDI-TOF MS results were successfully confirmed with 16S rRNA gene. Collectively, these data provide a seminal understanding of microbial populations in contaminated groundwater overlying anthropogenic activities like oil and gas development.” (emphasis added)

Study 3 opens almost the exact same way, mentioning unconventional development as the possible cause of contamination in the first few sentences of the Study 3 abstract:

Groundwater quality and quantity is of extreme importance as it is a source of drinking water in the United States. One major concern has emerged due to the possible contamination of groundwater from unconventional oil and natural gas extraction activities. Recent studies have been performed to understand if these activities are causing groundwater contamination, particularly with respect to exogenous hydrocarbons and volatile organic compounds. (emphasis added)

Even the press release from UTA – Three Studies from UTA’s CLEAR Lab Detect Harmful Pathogenic Bacteria in Texas Groundwater Near Natural Gas Extraction Sites – drives home a potential link between oil and gas development and groundwater contamination, without a single mention of other possible factors of contamination. As the very first paragraph of the press release states:

“Three new research studies from the University of Texas at Arlington have found harmful pathogenic bacteria in Texas groundwater near unconventional natural gas extraction sites.”

Choosing to focus on oil and gas development as the most likely source of water contamination in these studies is confusing not only because of the multiple statements by authors indicating no definite link between the UD and water contamination, but also because water analysis results used to support a link to UD activities and contamination could also be attributed to other factors.

For example, in Study 2, samples from the Barnett Shale are examined with volatile organic compounds (VOCs) and synthetic volatile organic compounds (SVOCs) included in the results table. These VOCs and SVOCs, the report notes, were “selected for analysis from a pertinent list of compounds commonly used in hydraulic fracturing.” With one of these compounds, acetonitrile, found in several of the samples, a reader would infer from the author’s presentation that oil and gas activities were the source of water contamination. However, acetonitrile is also used in agriculture for the production of pesticides, and can be found in automobile exhaust and exhaust from manufacturing facilities.

In other words, the presence of this compound does not necessary indict oil and gas development as the source of contamination as the study would lead you to believe. 

Conclusion

Overall, these studies support the conclusion reached by other scientific work: shale development is not linked to systematic groundwater contamination. To their credit, the authors say so themselves.
However, presenting these findings as suggestive of contamination from oil and gas development fails to provide the whole picture. Greater context on other potential sources of contamination would strengthen the study’s findings considerably – not because one source is more or less likely to be the culprit, but because the findings do not justify the intense focus on only one pathway (i.e. oil and gas).