The philosophy that University of Colorado research associate E. Michael Thurman applies to scientific research, he says, is: “You can sort the error from the truth if you work hard enough.”
This week, that task became far more difficult as Thurman and his research associates came under fire for apparently declaring the fluid used in hydraulic fracturing operations to be harmless.
But it wasn’t true. The researchers never said anything like that, nor did they intend to.
Like the children’s game of telephone, as word spread from one mouth to the next, the truth got so mired in errors it was nearly invisible by the end.
So how did a study designed to analyze traceable components of fracking fluid so potential contamination in groundwater could be identified get transformed into a headline that declared fracking fluid safe?
The answer is poor communication and bad journalism.
It started with an unclear press release from the University of Colorado with a title that declared “Major class of fracking chemicals no more toxic than common household substances.”
That choice seems to have bred confusion.
“There were quite a few news outlets that missed the distinction that we were trying to make about it being one class of chemicals — it’s an important class of chemicals, but it’s not all the chemicals, and basically [headlines] left the impression that all chemicals in fracking fluid were safe,” says Laura Snider, University of Colorado media relations staff member and author of the original press release.
“Ultimately, I feel like the news release is accurate. I feel like if I had to write again, I probably would make some things more clear or make some things that were down lower up higher, I guess. But I feel like it’s all in there.”
She admits her choice to place information about the toxicity of the chemicals the researchers had identified as a portion of the fracking fluid at the top of the release stemmed from an interest in making the science more compelling to readers far outside that field of research.
“The challenge that we always have writing about science is how can we make this something that people can understand or that they think is valuable or sort of has an impact,” she says.
In the days following the press release, and amid a deluge of phone calls Thurman and his research partner Imma Ferrer received about the study, Snider received one from the Society of Environmental Journalists questioning her approach.
“I think some professors at different universities were like, ‘We’re going to use this as an example for our class of what not to do, in the sense of — here’s the study, here’s the press release, here’s the actual coverage. Now where is the breakdown?’” Snider says.
She gave them a copy of the news release and says they told her they, like she, thought it was clear, but that there was a communication gap somewhere.
Nearly a week after CU’s press release, Colorado State University’s news website, Source, worked with the researchers at that university who had also been involved in Thurman’s study, to put out a story on the research. CSU’s reporting on the research puts at the forefront the statement that the purpose of the research was to provide a technique and database to use to determine if chemical compounds found in groundwater might have come from fracking fluid operations. It’s a stark contrast, and one with an obvious symmetry to the findings in the paper.
Unfortunately, the media buzz had already run away with CU’s very different story.
The Daily Camera’s front page headline the day following the CU press release declared, “CU study: Fracking fluid toxicity minimal: Researchers say liquid no more dangerous than household products.” This headline and the first paragraphs of the story omit the fact that the researchers studied only about onefifth of the chemicals in the fracking fluid samples they examined.
The Camera story concluded under a subhead “‘Facts aren’t so scary,’” with a declaration that the research shows that hydraulic fracturing is a fundamentally safe technology. That declaration came from the Independent Petroleum Association of America’s research, education and public outreach campaign tool, the website Energy in Depth.
The Colorado Oil and Gas Association also seized on the Camera’s misleading headline when it tweeted that the research adds evidence to COGA’s previous statements that fracking fluid toxicity is “minimal,” and “no more toxic than common household products.”
The official statement on Thurman’s research from Doug Flanders, director of policy and external affairs for the Colorado Oil and Gas Association, was: “We welcome and embrace sound science, thorough studies, and continued transparency. For Colorado families, this should again give comfort that oil and gas development is being conducted responsibly. It’s critical to note that in Colorado any concerned resident can already learn exactly what’s in fracking fluid, thanks to the state’s first-of-itskind disclosure rules. This is another example of how Colorado’s tough regulations on oil and gas development are working.”
Such reponses make one wonder how many of these people even read the study, which was published in the academic journal Analytical Chemistry in August and is freely available online.
What the research really said
The paper, “Analysis of Hydraulic Fracturing Flowback and Produced Waters Using Accurate Mass” was the first of several papers Thurman says he hopes to release on the subject of fracking fluid.
Thurman works in the Center for Environmental Mass Spectrometry at the University of Colorado’s Department of Environmental Engineering as does his fellow researcher, Ferrer. They aren’t toxicologists, he stresses, they’re environmental chemists.
“We weren’t out to find whether the [chemicals] were toxic or not,” he says. “In the press release, [Snider] wrote the word toxic in there, but that wasn’t our goal and we didn’t do any tests to determine toxicity. … What she wrote was not incorrect, but that wasn’t the purpose of the study.”
During their research Thurman and Ferrer ran the fluid through a mass spectrometer to examine its chemical composition, and identified the use of two surfactants — chemical lubricants used in household products including detergents, shampoos and soaps that reduce the tension between water and oil — used in fracking fluid. It’s a rare precision in an industry often shrouded with proprietary secrets.
Thurman’s focus was on identifying what percentage of the chemicals in flowback water from fracking operations are surfactants and to identify what type of surfactants were being used and produce what is essentially a fingerprint of those chemicals that can be traced.
Surfactants are known ingredients for hydraulic fracturing jobs, and are listed on FracFocus’s Chemical Disclosure Registry website. Thurman and Ferrer had previously researched surfactants, and had the instruments necessary to study them.
“We thought, this is a class of compounds that we can analyze, so let’s find out what percentage of the organic materials that they add to the well are coming out of the surfactants. So that was goal one,” he says. “And goal two was could any of these surfactants be used to fingerprint the individual wells themselves.”
The results provide a tool other labs can use to determine if groundwater has been contaminated by fracking operations nearby and, if there are multiple companies operating in the area, possibly identify which company’s frack fluid had entered the groundwater.
“We’d have to have a sample of what went down [the fracked well], but if you had that, I would guarantee you we could distinguish [between them],” he says. If someone wants to know if their well has been contaminated, they could answer that question.
“That, to me, is the valuable component [of the research],” Thurman says. “The toxicity question revolves around a lot of other compounds that are in there. There’s biocides added and there’s other compounds that maybe we haven’t figured out yet, and who knows about how toxic they might be.”
Biocides are known toxins that can be fatal at certain exposure levels. And numerous other ingredients in fracking fluid are suspected carcinogens.
Focusing on the toxicity of the chemicals, as both the CU press release and the Camera story did, both mischaracterizes the goals of the study and fails to acknowledge the limitation its lead researcher has openly pointed out. The study is based on a sample size of eight, wasn’t focused on toxicity and did not look at 80 percent of the ingredients found in fracking fluid.
Since the paper was completed roughly six months ago, Thurman has been able to run an additional eight samples. That’s still not much, he says.
“When you’ve only run less than 20 samples and there’s 7,000 wells just in Colorado alone — we don’t have a clue as to what’s really out there,” he says. “That would be like guessing who’s going to be president by asking eight people. I mean it’s not a very good sample set.”
The samples he obtained, including the eight in the study, were collected from five states (including Colorado) by way of colleagues. He also was given a standards sample and two of the flowback water samples by Multi-Chem, a Halliburton Service. Some came from colleagues at Colorado State University.
In exchange for the samples, Multi- Chem asked to read the paper before it was published. Their only objection was to the inclusion of a name of a specific surfactant mixture that included both types of surfactants, which they said was proprietary and asked that the name be removed.
The surfactants Thurman and his fellow researchers were studying make up only 20 percent of the fracking fluid — 80 percent of it is other stuff, perhaps largely water and sand, but other chemical components as well. Numerous sources claim that 350 to 700 chemicals are generally used in most fracking fluids.
“The other 80 percent, we don’t know yet. Some of it could be natural, harmless compounds, for example, acetic acid, which is vinegar, has been reported to be a major component coming back from the wells,” he says. “There are trace levels of other compounds that are known toxins, things called biocides that are added to kill bacteria so they don’t plug the wells, there’s some of that being put in the well. We’ve not worked on those, but that’s the next topic, probably, we will take a look for those. In fact we’ve already found some, but that’s the topic of a new paper.”
The surfactants Thurman found in the fracking fluid samples he obtained are, indeed, used in common household items — dish soap and hand lotion among them. One even has medical uses as a laxative and is a food additive approved by the Food and Drug Administration.
The compounds he studied, polyethylene glycols and linear alkyl ethoxylates, are so well known that when he approached Analytical Chemistry with the paper, they nearly turned it down.
“Analytical Chemistry is the premiere journal in the world on this topic and they only do new and interesting things, so if your work has been done somewhere else, they’ll reject you without reading the paper,’” Thurman says. “So you have to make a really strong case that what you’ve done is something new and exciting, and I had a little bit of a struggle initially because the two classes of compounds are well known … but no one’s measured them in groundwater, and especially for fracking water, so that’s what was new and exciting.”
Thurman and Ferrer did check for and failed to find a compound that is known to be an endocrine disruptor that has been listed in previous literature as used in fracking fluid in the samples they were given. Part of that could come down to the sample size — he started with just eight samples, and has since done eight more.
“It’s possible that had we done 100 samples, maybe there would have been some detections,” he says.
Contributions to the paper also came from Colorado State University’s Jens Bloetevogel and Thomas Borch.
“Now that we can identify these complex surfactant mixtures, we can better understand how these compounds move and under what conditions they may change into something else and how that may affect our environment,” says Borch, a CSU professor of environmental soil chemistry, according to Source. He and Blotevogel, an environmental engineering research professor, said the technique is a move toward better understanding the effect hydraulic fracturing may have on groundwater, as well as what happens to surfactants when they’re injected into the ground in fracking operations. They suggested the fingerprinting technique developed with this research could be used to help with monitoring water quality and perhaps remediation.
A small portion of the little financial support that went into this research came through Colorado State University and the Borch-Hoppess Fund for Environmental Contaminant Research. Thurman says that organization may have received grants from Halliburton, but if any of that money came to Thurman, it was indirect. He received no direct funding from oil and gas companies, he says.
We still don’t know exactly what’s in frack fluid — companies publish only a generic list of ingredients on the FracFocus website despite the claims to the contrary by COGA.
“It would be like if you were baking bread and you had a nice recipe, and you’d say ‘Well, I have wheat and I have sugar,’” says Thurman. “But you’re not telling us how much so we can’t reproduce your recipe, and also you’re using general terms. Maybe you use a special kind of flour. Well that’s what every body does with these wells,” he says.
With surfactants, the information made publicly available based on what environmental legislation requires is often generic enough that it could apply to a hundred compounds. Companies can always change the formula, too.
Among the compounds used in fracking fluid, according to data publicly released, are compounds that are known to be hazardous. How much is being used, and what happens to them when they’re mixed into that recipe, the amounts in flowback or the way they mix in the heat and pressure at 10,000 feet underground, where fracking operations take place, or what happens as they may encounter or pick up radioactive material or brine on the way back to the surface are still unknown.
Using Thurman’s technique to identify everything in fracking fluid would take years and cost a fortune, he says.
But Thurman’s work does identify chemicals that can be used as tracers — and provides information that could be used by private labs to identify fracking fluid. And that could be useful in instances of groundwater contamination.
“A lot of people are concerned that their water wells could be impacted. … Let’s say there’s some oil in their water and they think it’s coming from fracking. Could we detect the fracking fluid itself that went down the well? The answer to that is, we think we can now with this class of compounds and the reason is the levels are high enough that we can detect them,” he says. “They’re transported in groundwater. They’re not removed easily. They don’t degrade easily.”
All of that means, their chemistry endures in a way that makes them possible to trace and identify. They won’t change or absorb into the surfaces around them.
On the question of this ability to trace well contamination sources, COGA’s Flanders said via email, “With the Colorado water pre- and post-baseline testing rules and strict testing protocols, we already know hydraulic fracturing is not impacting drinking water.”
Yet public concern about groundwater contamination persists.
“Sometimes, the general public is aware of problems before the science is there,” Thurman says. “This is one of the first papers published on the flowback water … but people have been worried about it for years.”
With the constant presence of fracking on the news, the barrage of advertisements from oil and gas companies declaring it safe, the communities across the country voting on whether to allow it in their communities, Thurman says, it looked like a topic worth studying.
“It’s probably the most important political environmental topic in the U.S. right now,” he says. “So it seemed like a really important environmental topic for us to be involved in. … It doesn’t matter if you’re a Democrat or a Republican, you can’t ignore the fact that there’s that much oil that’s available with these new techniques, but it has to be done responsibly. What we’re trying to do is contribute just a little bit to that responsible recovery.”
As his research continues, he’ll be looking for additional samples of flowback water. For a fee, the lab could conceivably test well samples if people want to see if these surfactants are appearing in their well water near fracking operations. Running individual samples is tough, though, and a group or community-wide effort would be easier to navigate, were someone interested to pursue that option. Private labs equipped to test for these compounds could also now answer that question.
Among the areas in need of more research is whether surfactants — which, again, are lubricants — being pumped into the ground in the area of old geologic faults might be contributing to the earthquakes recently reported near injection wells that take flowback water from fracking. Knowing that the flowback water is likely to contain surfactants that were used to lubricate the rock to allow for the extraction of oil raises some questions of its own.
“So you’re pumping this down into old faults that haven’t moved, and you’re helping to lubricate them a little bit — that may not be a good idea,” Thurman says.
His next step will be to look at these compounds they’ve now analyzed in the environment, investigating questions like what happens to them in the heat and the pressure found 10,000 feet under ground during fracking operations. He’ll also begin studying the biocides used to kill off microbial life that might clog the pipes for oil and gas — some of the toxins that do exist in fracking fluid.
Science, he says, works a little bit like a set of puzzle pieces that come with pieces that don’t fit in the puzzle mixed in among them, and no image to clarify what you’re building. There are a lot of pieces still to place when it comes to fracking.
“That’s what science often does. The data is coming at you, the information, some of it is relevant, and some of it has nothing to do with it at all, and your job is to figure it out,” Thurman says. “So I walk along and I say, ‘What is the one thread of truth that’s out there?’ And you start pulling on that thread and then pretty soon, boom, there it is. There’s the story. That’s research.
“There’s a lot of similarity to good journalism, investigative journalism, and research. It’s really the same thing, and you’re searching for the truth, trying to dig it out, and the subtleties of truth.”