University of Vermont researchers have released a new study in the journal Water Research Resources that argues fractures generated during hydraulic fracturing could intersect with abandoned or existing wells if such activity were to occur in the Marcellus Shale underlying New York State.  But what you probably didn’t see in the headlines on this study is the fact that the authors concluded that the possibility of such intersection is very low and that, were it to occur, it would not necessarily mean that fluids or gasses could make their way to the surface.

Here are four things you need to know about the study:

1. The researchers concluded that the likelihood of well intersection was very low

The researchers’ statistical modeling indicated that the probability of a new well intersecting with an existing well was zero-to-3.45 percent.

In fact, the researchers found that some regions in New York underlain by the Marcellus Shale “have probabilities on the order of 1-2% of encountering existing wells, while the vast majority of regions have much lower probabilities.”  (Emphasis added). The probability calculation is evaluated for each subdivided region or representative area of the region. For example, the researchers found that the average probability is approximately 0.025 percent for one representative area of 12,000 square feet.  The figures vary based on the density of wells assumed to be in a given area.

It is worth noting that this study is necessarily hypothetical as there is no fracking in New York due to Gov. Andrew Cuomo’s ill-informed  ban on the practice, which made unethical use of the peer-review process and ignored the scientific consensus that fracking is a safe technology with manageable risks. This fact that has not prevented New York from importing natural gas into the state nor has it stopped New York politicians from touting the climate benefits of Marcellus gas.

2. Researchers find well integrity failure is needed for a pathway to the surface to occur; well failure rates are less than one percent of all wells drilled

The authors were careful to note that even if well-intersection occurs that does not necessarily mean that fluids or gas will reach the surface.  From the text:

Encountering an existing well does not guarantee that a wellbore is capable of providing a pathway to the surface. (Emphasis added)

Further:

Potential pathways include: cracks between cement and surrounding rock formations, spaces between the cement and well casing, voids created by well casing corrosion, and pathways through the interior of the well itself.  Pathway development is dependent on whether a wellbore has failed, and failure is dependent on the quality of construction, lifetime monitoring, and the care taken during shut-in [Wojtanowicz, 2008; Bachu and Watson, 2009; Watson and Bachu, 2009]. (Emphasis added)

The “potential pathways” noted by the authors are already well understood and heavily regulated, which is why the U.S. has extraordinarily low well failure rates.  In fact, data from an Associated Press investigation shows that the well failure rate for oil and natural gas wells drilled in Pennsylvania is approximately one third of one-percent (0.33 percent).  In Texas and Ohio, a 2011 Ground Water protection study found a well failure rate of less than 0.01 percent and 0.03 percent, respectively.

The concern that fracking can cause a pathway for methane leakage has also already been the subject of several rigorous studies, all of which indicate that methane emissions have dramatically decreased as natural gas production has soared.  To this end, the U.S. Environmental Protection Agency (EPA) has determined that U.S. methane emissions fell 16.9 percent since 1990 and the Intergovernmental Panel on Climate Change (IPCC) has noted that as an effect of more power plants converting to natural gas, “recent lifecycle assessments indicate that specific GHG emissions are reduced by one half.”

Current estimates for methane leakage are very low. EPA studies estimate that the leakage rate for natural gas systems is about 1.5 percent while numerous studies conducted by the Environmental Defense Fund have shown leakage rates are well below 3 percent; the rate which scientists claim methane leakage from fracking would erase the benefits of natural gas.

3. The authors acknowledge the limitations of their methodology

In seeking to determine the potential for well intersection the researchers created a data set of known and unknown wells and plugged it in to a statistical model.  The data set was constructed from a series of assumptions and extrapolations; a method that has limitations, which the authors recognize:

The large number of unaccounted for wells within New York State means that determining exactly where new hydraulic fracturing (HF) wells will intersect existing wells is not possible and necessitates a probabilistic approach to existing well locations and depths.  This requires several assumptions to be made with regard to existing wells and fracture propagation. (Emphasis added)

These assumptions include the very notion that the known wells are a subset of all wells in the region and that wells located in the same area target the same formations.

4. The researchers draw modest conclusions

The authors appropriately positioned their study as a complement to the existing body of research; citing studies that have shown that fluid migration (gas, brine, and fracturing fluid) would not occur without a pathway:

The large distance through overlying rock formations to shallow aquifers makes fluid migration unlikely without the presence of discontinuities to provide pathways through the rock [Leff, 2011; Flewelling et al., 2013]…

Upward propagation through natural fractures… is deemed impossible by Engelder et al. [2014]. (Emphasis added)

They also cite studies that indicate even when a pathway is present, fluids are still unlikely to migrate to shallow aquifers:

While it may be possible for liquids to migrate upward, the conditions required to reach shallow aquifers or the surface seem unlikely to occur. Flewelling and Sharma [2014] provide a good analysis of the conditions necessary to mobilize fracturing fluid or brine.  (Emphasis added)

Due to the limitations conferred by their methods, as well as the very low probability of well communication identified in their conclusion, the researchers did not overstate the meaning of their findings.  They concluded:

The probability of encountering a well is the first step in assessing the risk of a hydraulically fractured well communicating with shallow aquifers, and places an upper bound on that risk.  Further work on assessing the probability of an encountered well providing a pathway needs to be done to determine the risk that fracturing poses to shallow aquifers. (Emphasis added)

Therefore, this study’s conclusions are not what some media outlets would have you believe.  In fact, the researchers write:

Since the integrity of the wells and borings are not considered in this paper, the results found do   not represent the actual risk associated with hydraulic fracturing by means of communication with exiting wellbores, but instead place an upper bound on that estimation.

Activist and Media Reaction

Activists and the media have pounced on this study to claim that it demonstrates significant danger of excessive methane leakage to the surface, however as noted above, this is false.

Take this lede from a Reuters article about the study:

Hydraulic fracturing can cause nearby abandoned oil wells to leak methane, according to a study published on Tuesday in the peer-reviewed Water Resources Research journal, marking a potentially large source of unrecorded greenhouse gas emissions. (Emphasis added)

This is not only a complete misrepresentation of the study’s conclusions, but it also fails to grasp what the researchers were actually examining.

The Sierra Club and Frack Action were quick to draw attention to this Reuters article and trumpet to their Twitter followers that fracking will cause methane leakage.   Despite the fact that this study did not—at any point—conclude that intersecting with an existing well was likely or that methane leakage would necessarily result if an existing (active or abandoned) well was hit.

Conclusion

Despite misleading media accounts and tweets, this study concludes simply that there is a low probability that newly drilled could intersect with existing wells. This study positions itself as a part of a body of research that validates what experts consistently report: fluid or gas migration is unlikely at best, and, as scientists consistently point out, fracking is a well-understood technology that has been used for many decades without the environmental problems imagined by activists.  This includes no evidence of fluids or gas “migrating” through thousands of feet of impermeable rock.

As was repeatedly noted throughout the study, “due diligence by oil and gas companies during construction in identifying all nearby wells will have the greatest effect in reducing the probability of interwellbore communication.” Thankfully, oil and gas companies have excellent track records of compliance with industry best practices, laws, and increasingly strict regulations, including regulations related to well integrity, water quality, and air quality.