A recent study has found that, under certain conditions, the chemical-laced water used in hydraulic fracturing can migrate through fractures and faults up to overlying aquifers in as little as tens of years.
The study, done by hydrogeologist Dr. Tom Myers and published in the peer-reviewed Ground Water, raises renewed questions about the potential for hydraulic fracturing to fundamentally alter shale rock formations and the hydrogeologic cycle in ways that could affect freshwater drinking supplies.
To reach gas in shale rock, a well bore is sunk thousands of feet into the earth, passing through different strata, including freshwater aquifers, sandstone and siltstone, down to the shale. Shale can include remains of the breakdown of trapped organisms that settled out with silt and clay-sized particles, making up the oil and gas. To access these fossil fuels, thousands of gallons of water with chemicals and silica particles (sand) is blasted horizontally into the shale under pressure is so high it could chip the paint off a car, in a process called hydraulic fracturing, or fracking.
According to the oil/gas industry's FracFocus, the process works "to create or restore small fractures in a formation" that become "paths that increase the rate at which fluids can be produced from the reservoir formations." To protect overlying aquifers, "steel surface or intermediate casings are inserted into the well to depths of between 1,000 and 4,000 feet. The space between these casing 'strings' and the drilled hole (wellbore) ... is filled with cement."
There are concerns that improperly sealed or unsealed abandoned drilling wells and naturally occurring fractures and faults will offer a path to connect gas bearing shale layers with overlying freshwater aquifers.
A common response by the oil/gas industry to this concern is that upward migration of deep, thermogenic methane gas and chemicals used in the fracturing process is impossible, due to layers and layers of impermeable rock that prevent such movement.
The problem, some hydrogeologists say, is the industry is offering little to no data to support this argument. While the ability of shale to transmit fluids (its permeability) is about 1,000 times less than that of sandstone aquifers, hydrogeologists say the entire purpose of hydraulic fracturing is to increase the permeability of shale rock to allow the "trapped" methane gas to leak out, mainly into the more permeable overlying rock layers. Pre-existing fractures provide such a pathway.
Paul Rubin, a hydrogeologist with the environmental consulting firm HydroQuest, said savvy companies will target such fractures: "[oil/gas companies] are integrating pre-existing fractures. When they put in horizontals [drilling wells], they go perpendicular to those fractures to maximize the number of fractures they go through and the gas potential."
Myers said the low porosity of shale itself could propel movement of fluids: "I don't know where [oil/gas companies] think all that water would stay underground. I mean, the pore space, the porosity is not there for it to be retained, you're putting a lot of pressure down there and it's going to move somewhere."
Up to a third of the fluid used in hydraulic fracturing will resurface, as well as naturally occurring and extremely salty brine, or "produced water." The high amounts of the resulting wastewater from the fracturing process has raised its own challenges around disposal and treatment, as well as the potential for water contamination from spills.
But it is the water underground that Myers focuses on in the Ground Water study. Brine has been found more than a thousand meters above its evaporative source, suggesting evidence of upward movement. The question, Myers said, is how quickly this movement occurs and how the fracking process might affect the rate: "Fracking shale moves water that could meet up with naturally occurring fractures in sandstone and result in faster movement to the surface [aquifers]," Myers said.
To determine this, Myers created five conceptual models to test different parameters and the rate of flow from the Marcellus shale to the level of aquifers.
The first model offers a baseline of natural upward flow of fluids from shale to the surface without hydraulic fracturing of the shale, which produced a rate of around tens of thousands of years, consistent with what is observed in the area with the natural, upward movement of brine. The second model added a fracture from the shale to the surface and found that this sped up the flow rate of fluids tenfold, suggesting that fractures and faults can offer a shortcut for the upward fluid movement.
The third and fourth models looked at how long it takes for the hydrogeologic system to come to equilibrium after hydraulic fracturing - model three without a fault and model four with a fault. Model five simulated the actual injection of fluid into fractured shale with and without a fault.
Under certain circumstances, Myers found that the pressure induced by hydraulic fracturing combined with the changes to the fracked shale and the presence of a fault could mean that fluids migrate upward from shale to aquifers in as little as hundreds to tens of years.
Myers summarized the results: "You change the properties of the shale, you change the flow of the system. If the fractured shale between each of the well bores almost connects, you end up with a system of several orders of higher conductivity. You end up increasing the flow rate."
The gas industry has been quick to point out that the study was commissioned in part by Catskill Mountainkeeper, a New York-based environmental group that opposes hydraulic fracturing.
Some geologists have challenged not the funding, but the methods of Myers's study, such as Terry Engelder, a professor of geosciences at Penn State. According to NPR's "This American Life," Engelder rose to prominence with the oil/gas industry after releasing his high estimations of the amount of gas in the Marcellus shale. Engelder argues that most of the overburden in Myers's model is permeable sandstone, while the actual overburden contains 90 percent shale and only 10 percent sandstone. "If the sandstone were replaced by shale within Myers's model, the time frame required for water movement to the aquifer would increase to 100,000 years," Engelder told Energy Wire.
Myers disputed Engelder's claim: "His statement does not agree with descriptions in the geology section of the New York Draft Supplemental Generic Environmental Impact Statement, which describes it as a mixture of shale and sandstone. I considered a broad range of hydraulic parameters to encompass the full range of possibilities. [Engelder] is focusing on the very tight end of the potential range in hydraulic properties."
Myers said that rather than just rejecting his work, drilling companies should work with him on collecting data and improving the studies: "Don't just say that I'm using the wrong parameters - collect some data to show what the parameters should be and let's do future modeling and future data collection."
Other hydrogeologists like Rubin are concerned about not just fractures and faults, but the potential for water contamination through drilling well casing failures. "It is not if, but when it will happen and it will happen," Rubin said. His studies have focused on documented cases of cement and steel casing problems and the relatively short life spans (maximum 100 years) of the casings compared with the million year life span of the aquifers they are supposed to protect. "[Well] casing will degrade and even if that does not end up being a problem, the fracturing process itself will result in cracking of the cement sheath. Also, a lot of these wells are in seismically active areas and ground motion and shaking will themselves crack the sheath."
The issue is gaining renewed importance as New York Gov. Andrew Cuomo weighs whether to allow for hydraulic fracturing in the state.
There is currently limited federal research to settle the issue. The Environmental Protection Agency (EPA) began preparing its report on hydraulic fracturing and drinking water in 1999 after an Alabama court legally compelled the agency to investigate reports of fracturing-related water contamination in the area. The EPA focused its study solely on coalbed methane deposits, which are typically more shallow than shale gas deposits, and in 2004 concluded "there was little to no risk of fracturing fluid contaminating underground sources of drinking water during hydraulic fracturing of coalbed methane production wells."
A few months after the report's release, the 2005 Energy Policy Act was passed, exempting the practice of fracking from the federal Safe Drinking Water Act, based in part on the EPA study. By default, the responsibility fell to individual states.
ProPublica later reported that buried within the 424-page EPA report were statements explaining that the fracking fluids migrated unpredictably - through different rock layers and to greater distances than previously thought - in as many as half the cases studied in the United States. It found that as much as a third of injected fluids, benzene in particular, remained in the ground after drilling and is "likely to be transported by groundwater."
In public comments on New York's Environmental Impact Statement concerning hydraulic fracturing in the state, concerns were raised that the state's drinking water standards are now outdated, as they do not account for the novel mix of various chemicals used in hydraulic fracturing. Without updating the standards, oil/gas companies can argue that they are within drinking water standards, because those standards have not been developed.
In several states where hydraulic fracturing is occurring, many of the chemicals used and their concentrations remain unknown, protected through legislation as proprietary trade secrets. The legislation was later traced to a "model bill" sponsored by ExxonMobil and promoted through the corporate-funded American Legislative Exchange Council (ALEC), according to The New York Times.
Myers cautioned that until more data is collected on the way that hydraulic fracturing affects the hydrogeologic cycle, the process should not occur in sensitive or highly populated areas.
Rubin thinks that what is already known about well casing failures is enough to put a hold on unconventional drilling: "Why would anyone risk our aquifers for a few years of profits?"