U-turn prof finds his fracking fears are avoidable
- Ivo Vegter
- 11 Feb 2014 (South Africa)
“A Free State professor says there is a high risk that hydraulic fracturing in the Karoo could cause the worst water pollution problem in the world.” Thus ran the lead paragraph of a Sunday Times story on 1 June 2012.
The professor, Gerrit van Tonder of the Institute for Groundwater Studies at the University of the Free State was, in his own words, “100% certain that there will be trouble”. More specifically, he told the media: “There is a high risk that fracking in the Karoo could lead to one of the biggest water pollution problems in the world.”
The claim is based on the theory that there exist briny aquifers, at the level of Ecca shale some 4,000 metres below ground level (mbgl), that are under artesian pressure. The flow gradient of deep groundwater is, therefore, upwards, and via preferential pathways – either along natural intrusions of dolerite or via defective well bores – this water can reach the surface. Deep brines are already contaminated with heavy metals and radioactive substances, and won't be improved by having fracturing fluid added to it.
Time is a healer, of course, and most harmful chemical substances are broken down, sequestered or diluted by chemical, physical and biological processes over time. This is why toxic brines at depth aren't already a widespread (albeit natural) environmental crisis.
However, Van Tonder claimed that the contamination of freshwater aquifers due to hydraulic fracturing could occur on a time-scale of years or decades, rather than centuries or millennia.
His evidence for this is that artesian water flowed from a long-abandoned oil and gas exploration well drilled in the 1960s. There is also, he told Julienne du Toit in these pages, some anecdotal evidence that a chemical used in drilling one of the wells reached a water well some 30km away in a matter of weeks.
To be fair, in private communication Van Tonder chose his words more carefully. For example, he distinguished clearly between “contamination” and “pollution”, noting that by the former, he means only that trace chemicals exist at a detectable level, while the latter implies levels that are unacceptably high. No newspaper stories I've seen made this distinction clear.
Even if they did, a lot of questions remain. Some of them I raised in columns published at the time. Are modern wells drilled and sleeved in the same way 1960s wells would have been constructed? Does the artesian water come from shale depth, or from shallower aquifers? Why do hot springs in the Karoo Basin all originate from much shallower aquifers, and would it be okay to fracture shale at depths greater than 1,500 mbgl, as exploration companies have said they intend? How reliable is the sole anecdotal case of pollution, and can such cases be prevented? If dolerite intrusions create preferential pathways for upward flow, why would defective well bores make the situation any worse than it already is? Are there technical solutions, such as self-healing cement, that would minimise the risk of well sleeve deterioration?
The trouble was that none of Van Tonder's anecdotal work and modelling was part of the formal scientific literature on shale gas exploration in South Africa. Scientists I approached were not prepared to comment on what they described as an unpublished “pet theory”. They expected publication in a “reputable journal”, adding that no South African journal would meet this standard. (Sorry, Water SA.)
At the time, Van Tonder claimed that the issue was too urgent to await scientific publication, since peer review could take as much as nine months. Fourteen months later, in July 2013, he sent me a draft paper, and told me that he'd submit it to an unspecified US journal for publication. It was co-authored with Fanie de Lange, Gideon Steyl and Danie Vermeulen, and titled Potential Impacts of Fracking on Groundwater in the Karoo Basin of South Africa.
In November 2013, he did submit a paper for publication. It appeared on 5 February 2014, three months later. So after waiting almost two years, we finally have a published Van Tonder paper. Here it is.
It has a new title, Stochastic Risk and Uncertainty Analysis for Shale Gas Extraction in the Karoo Basin of South Africa. It has a new co-author, Abdon Atangana, whose speciality appears to be numerical methods and modelling. It has a new publisher, the journal Abstract and Applied Analysis, a product of the Hindawi Publishing Corporation of Egypt.
After notifying me of his newly-published paper, Van Tonder told me that his proposal for plugging abandoned wells, and drilling three monitoring wells 100m deep around the gas well site, would mitigate the risk. “Then I will be happy,” he said.
So, how did we get from the fears peddled to the media in 2012, about being “100% certain” of “the worst groundwater pollution problem in the world”, to being happy? That seems an important question to answer. The alarmist headlines of 2012 are engraved on the minds and posters of anti-fracking protesters country-wide. That horse has bolted.
If the 2014 paper supported the 2012 claims, it would be harder to berate the media for running with sensationalist fear-mongering based on simplistic and premature speculation. But it doesn't. The paper does little beyond present a model that demonstrates what might happen under certain assumptions. As models go, it may be excellent, but a model is only an idealised representation, and is only as good as its assumptions.
While we await formal scientific comment on the paper, let's consider some of its more obvious features.
The Hindawi journal is one of a fast-growing number of open-access academic journals that charge authors a fee for the service of peer-review and publication. In this case, the fee is $1,200.
In principle, open access is a great idea. It has long been a source of frustration to me, as a freelance journalist, that it is difficult to access scientific papers behind pay-walls when you don't work for an institution that has appropriate subscriptions. Even one-day access to single papers can run into hundreds of rands.
However, the author-pays model does have its detractors, because of obvious conflicts of interests. Some in the scientific community call them predatory, or the dark side of publishing. Hindawi has been described as dodgy, and its flagship journal has been the subject of controversy over fraudulent attempts to boost its standing, or impact factor. It appears to have taken corrective measures, has been removed from a widely-read blacklist, and does appear in a directory of nearly 10,000 open-access journals that claims to assesses them for quality.
Not being a scientist myself, I can only guess at Hindawi's true standing, but it seems fair to give it a pass.
Not so, say scientists I approached for comment on Van Tonder's paper. The two who got back to me before my deadline both asked that I withhold their names for professional reasons, though neither have ties to the oil and gas industry.
One claims never to have heard of the publication. He also noted that it is poorly edited for grammar, which I wasn't going to raise, but is most certainly true. The other said: “I wouldn’t be seen dead publishing in these fly by night journals. They keep asking me to be on editorial boards; look at the list of unknowns – this is the sort of open source publications anyone can get a paper in, and is doing harm to the quality science community.”
But, now that we finally have something concrete to consider, let us not be churlish, and try to evaluate the paper's content.
One point that raised an eyebrow at the outset was the citation of Osborn, Vengosh, Warner and Jackson, of Duke University. If you followed the shale gas story, you may recall it as the study that found higher levels of methane in selected water wells near gas drilling operations.
It was widely criticised for its small sample size and cherry-picking wells for testing. It found none of the tell-tale chemicals that could have acted as tracers to prove a link between shale gas wells and groundwater contamination. And while the authors themselves took to the newspapers to trumpet “strong evidence that shale drilling is risky”, this contradicted the paper's actual conclusion: “We found no evidence for contamination of drinking-water samples with deep saline brines or fracturing fluids.”
Subsequent studies, such as Molofsky et al, which sampled more than ten times as many wells, confirmed that stray methane in water wells was unrelated to shale gas operations.
It is unclear to me why a new scientific paper that purports to examine the contamination of freshwater aquifers by shale gas drilling would cite the Duke study, but not the more recent and more thorough Molofsky study.
Another new study, by Samuel Flewelling and Manu Sharma, entitled Constraints on Upward Migration of Hydraulic Fracturing Fluid and Brine, says that suggestions of high flow rates and short travel times for deep brines near hydraulic fracturing zones “contradicts the body of literature on the hydrology of sedimentary basins”. It singles out the work of Tom Myers, Daniel Rozell, Sheldon Reaven and Nathaniel Warner for criticism, all of whom Van Tonder relies upon.
Flewelling's paper concludes: “In sum, rapid upward migration of [fracking] fluid or brine via bedrock would require the co-occurrence of upward head gradients and high bedrock permeabilities. As we discuss in this article, these two conditions are mutually exclusive, indicating that widespread and rapid upward migration of [fracking] fluid and brine through bedrock is not physically plausible.”
Clearly, there is some disagreement here. My scientific sources were again more scathing about it than I have any right to be as a mere journalist. Asked whether it establishes the claims made to the media in 2012 that freshwater aquifer pollution in the Karoo is inevitable because of artesian brine aquifers at Ecca/Dwyka depths, one said, “No it does not establish any claim to that effect.”
The other elaborated: “[Van Tonder's paper] is so ‘opaque’ as to be just about meaningless to any of the laypeople and scientists currently interested in the effects of fracking in the Karoo. I presume that ‘sampling’ in the context of the paper is a statistical term and bears no relevance to water sampling. The authors acknowledge the data gaps and uncertainties and I think that this is just an academic and premature exercise. There’s nothing new here – just a lot of mathematical/statistical manipulations to try and put numbers to scenarios that we are already aware of and are striving towards gathering data to be able to analyse and predict realistically.”
He added: “I won’t be spending any more time trying to make sense of it and suggest that the local scientific community won’t take much notice of it except as another esoteric and premature analysis.”
But let us suppose that I, my sources, and all the scientists I cited are wrong. Let us suppose that it is plausible that hydraulic fracturing chemicals from deep shales can contaminate shallow groundwater in high enough concentrations and over short enough time spans to cause significant concern.
What does Van Tonder's published, peer-reviewed paper have to say about that? It concludes: “The study reveals that, in the case of the Karoo, the idea of fracking will be successful if and only if a well and the entire fracked reservoir are plugged with, for example, cement, otherwise many aquifers in the Karoo will be polluted.”
So, science, having speculated about a potential problem, has also provided us with a fairly simple preventative measure. And since Van Tonder once again serves on the Karoo Water Expert Group, it seems likely that such a solution will be included in the final regulations.
That should make everyone happy, but don't expect the activists or the sensationalist media to tell you this. Now, if you'll excuse me, I need to go and check if my sources have unholy ties to the cement industry. DM
Note: Daily Maverick has confirmed the identities of the anonymous sources quoted in this article, and is satisfied with their credentials as independent scientific experts. -- Ed.
Reader notice: Our comments service provider, Civil Comments, has stopped operating and will terminate services on 20th Dec 2017. As a result, we will be searching for another platform for our readers. We aim to have this done with the launch of our new site in early 2018 and apologise for the inconvenience.