Correlation does not imply causation; what are the factual risks of hydraulic fracturing on our drinking water supply?

“Throughout the history of literature, the guy who poisons the well has been the worst of all villains” – Author unknown

The midterm of only my first semester as a PhD student just recently passed and I am already wondering if hydraulic fracturing research has left me one dimensional, leaving me with nothing interesting to talk about beyond this subject.  I am clearly now the one to avoid during happy hour conversations. To add to this life consuming subject, I have been attending the Fracking Sense Series, organized by the Center for the American West.  Last week’s speaker was George King of Apache Corporation.  His topic was Hydraulic Fracturing and Well Developments: What are the factual risks? with a focus on water quality.

Water quality concerns associated with the hydraulic fracturing process first entered my world the moment I arrived in West Virginia.  A golden algae bloom released toxins into Dunkard Creek, a meandering stream along the West Virginia-Pennsylvania border.  At the time this creek was one of the most ecologically diverse waterways in the region.

Dunkard Creek; a tributary of the Monongahelia River
Dunkard Creek; a tributary of the Monongahelia River

The toxins ultimately resulted in a massive fish kill of more than 160 species of fish, salamanders and endangered mussels.  While there were numerous coal companies historically discharging total dissolved solids (TDS) into the creek, fingers were pointing to the natural gas industry as they were the new kids on the block.  Thousands of citizen youtube videos implicating the industry flooded the internet at the time.  Apparently, these videos could only tell the story to the music of Creed, so if you are not a fan select mute before pressing play.

The contamination concern with the natural gas industry was not with the hydraulic fracturing process itself, but with cross contamination potential of hoses and tanks. If you remember from my first blog post, the flowback or produced water from the drilling operations is disposed of in tanks and trucks via hoses. Because there is “no manifest system for water withdrawal” and “there is some amount of water that gets left in the tank and hoses that then gets put into streams” according to EPA biologist Lou Reynolds, the potential of the golden algae originating from this process is high.

However, years after the devastating fish kill reasons behind the chemical changes in the creek remain a mystery and are still debated by the EPA, the industry and locals. Reynolds maintains that the fish kill may have ties to the Marcellus drilling, stating that “something has changed in the mine pools” since drilling in the area began. It could be that miners are digging deeper into the coal seam which “could result in their disposing of Marcellus water in their mine pools or taking in water into their treatment ponds”.  Or either industry could simply be ignoring regulations and disposing a greater amount of TDS into the streams. According to Reynolds, “any combinations of these are a possibility”.  Ultimately, Consol Energy Inc. and their Blacksville No.2 mine was held responsible and entered into a multimillion-dollar settlement. Despite paying fines and building a treatment plant, Consol maintains that they did not cause the fish kill and the EPA never assigned blame.  Perhaps coincidental, but shortly after the settlement, a local waste hauler was arrested and charged with improperly disposing of produced water into tributaries of Dunkard Creek.

This is the typical water quality story associated with the hydraulic fracturing industry; whispers of fault, correlation (not implying causation), and settlements without admissions of guilt.  When you add scientific disagreement and nondisclosure contamination court settlements, it is no surprise that public fear remains high. This is why George King’s talk was so important, as he was to provide evidenced based findings on the water quality impacts of hydraulic fracturing.

King’s presentation focused on well integrity.  He explained the drilling and geologic fracturing process and the multiple cement barriers that surround the well in order to isolate the casing from groundwater and prevent natural gas from leaking up around the outside of the pipe. The number of these barriers is based on the hazard risk to our drinking water supply.  Typically there are up to four barriers at the aquifer or fresh water level, two barriers above to the ground surface and below to mid-level, and only one at the deep gas extraction level.  The graphic below illustrates a typical well casing configuration.

Well casing through the water supply and gas reserves geologic formations
Well casing through the water supply and gas reserve geologic formations

According to his data, the risks of well leakages or failures are quite small.  Leakage rates of modern wells are less than 0.00005% per volume of gas produced and the chances of a well failure are one in one million, with nearly all of these occurrences in either older wells or in areas where oil and gas regulations pertaining to well construction were weak.  In fact, the risk of hydraulic fracturing was so minimal it did not even make the EPA’s top 20 list of groundwater pollutants.

Hydraulic Fracturing and Well Development (2)_Page_13
EPA’s list of the largest groundwater pollutants

Beyond lax construction standards and aging infrastructure, King noted that the biggest risks during the fracturing process were surface spills of additives and produced water, which are standard risks of many industrial processes and not unique to hydraulic fracturing.

So what about the toxic non-biodegradable fluid left in the ground after a well is “fracked”?  It is estimated that only 20 – 40% of the fracturing fluid is recovered in the flowback or produced water, the rest is absorbed by the shale formation that is isolated from the water table by a significantly thick geologic formation (note that these percentages vary depending on the source).  While domestic wells can sometimes reach up to 500 ft below the surface, drilling typically occurs 5,000–10,000 ft below the surface. This geologic barrier between the two is the same impermeable layer that has kept oil, gas, brine and other fluids trapped for millions of years.  This is why the risks of migration are considered minimal.  The graphic below provides a glimpse of the relationship between the above ground operations and the below ground geologic formations.  The caption provides an interactive link.

Hydraulic fracturing operations and geologic barriors
Hydraulic fracturing operations and geologic barriers

This piece of King’s presentation was the only part that he did not substantiate with actual data.  While the inability of fluid migration was once a settled science, there are now questions surrounding possible migration through faults, fissures and any other unknown irregularities in the geologic formation.

I left the Fracking Sense Series that night with greater confidence in well integrity, but with substantially more questions about fracturing fluid migration risks. I also found the EPA groundwater pollutant list that he presented troubling, even with the absence of oil and gas extraction. While spills (#4) and shallow well injection (#17) do not occur during the fracture itself, they are part of the process that is necessary to obtain natural gas by hydraulic fracturing and therefore should be considered an associated or indirect risk.

Ruling out well integrity, what stage of the natural gas supply chain poses the biggest risk to our drinking water supply?  Is it in the treatment of wastewater, migration of fracturing fluids remaining in the shale formation, or simply the lack of regulation that is the potential natural gas connection in the Dunkard Creek case? I will spend the remaining posts investigating these very issues.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s