Article courtesy of Scott K. Johnson | September 18, 2014 | Ars Technica | Shared as educational material
The primary public concern surrounding fracking—the fracturing of shale rock layers with hydraulic pressure to release the natural gas and oil they contain—has been the perceived risk to drinking water. After all, the water used to fracture the rock is laced with chemicals that enhance the process, and some of them are hazardous. While those chemicals haven’t really shown up in water wells, natural gas has. If natural gas isn’t identified and vented, it could collect in buildings and pose an explosion hazard—videos of garden hoses turned into flame-throwers have made the rounds.
But tying that natural gas to fracking projects isn’t as straight-forward as many assume since there are natural sources of methane as well. One group of researchers has been studying this question for several years, focusing on Pennsylvania, where the Marcellus Shale has been targeted by the natural gas industry. A controversial analysis the group performed concluded that natural gas in well water was more common near active natural gas production wells, indicating that much of the contamination was related to recent human activities rather than natural conditions.
The researchers also looked for hints of natural migration of fluids from the Marcellus Shale, which is deep underground, to the well water, which is taken from sources closer to the surface. By analyzing elements like chlorine and strontium, they identified the fingerprint of briney Marcellus fluid in some of the water wells, which pull from an aquifer where concentrations of those elements are much lower. They concluded that some of those fluids were present, casting doubt on the idea that the Marcellus Shale was too tight a seal to allow fluid to escape upward into drinking water. That work also indicated that some of the methane-contaminated wells seemed to be impacted by naturally occurring methane, but typically the ones close to natural gas production wells weren’t.
Searching for nobles
That same research group has published a new study in the journal PNAS looking at what the analysis of noble gas isotopes in water samples can add to our understanding. That analysis includes 133 wells in the same area of Pennsylvania as well as 20 wells in Texas, where the Barnett Shale is being fracked for natural gas.
Noble gases have the potential to tell us where the gas in water samples came from partly because they are chemically nonreactive, making it easier to tell what they’ve been doing underground. Some isotopes, like neon-20 and argon-36, will be common in water recently exposed to the atmosphere. Others, like helium-4, neon-21, and argon-40, are produced by radioactive decay and will predominate in deep, old, isolated groundwater.
The researchers think that these isotopic signatures can help differentiate between gas that migrated slowly upward from natural gas sources above the Marcellus and gas from the Marcellus that took a shortcut through a production well before escaping, for example. Some of their well samples—the same ones that seemed influenced by deep brines—bore the fingerprint of the natural process of slow gas migration.
But some other wells less than a kilometer from natural gas production sites told a different story. The noble gas isotopic signature in those samples has been interpreted to match the shortcut scenario. Four clusters of Pennsylvania wells seemed to contain Marcellus gas that leaked out of the production well and into the drinking water aquifer. Three clusters looked more like gas from a shallower rock layer that escaped upward along the outside of a production well, where cement doesn’t always perfectly fill the space around the well pipe. An eighth cluster involved Marcellus gas leaking upward in a different manner, likely due to a mechanical problem in a nearby production well.
For the Texas wells, five of the 20 had a human-caused signature, probably relating to shallower gas migrating along the outside of the production well.
The fingerprinting these researchers are attempting is undeniably tricky, but it once again points to the seal in and around natural gas wells as the primary vulnerability raising contamination risk. That has always been more plausible than having the fracturing of the shale rocks allow the free flow of contaminants into overlying drinking water aquifers—at least for geological configurations like the Marcellus Shale. The researchers conclude, “In our opinion, optimizing well integrity is a critical, feasible, and cost-effective way to reduce problems with drinking water contamination and to alleviate public concerns accompanying shale gas extraction.”
A separate study published in the Journal of Unconventional Oil and Gas Resources corroborates that general picture, although it challenges some interesting details. For those concerned about fracking fluid contaminating drinking water, the fact that less than half of that fluid often comes back up when the job is done doesn’t inspire confidence. So where does it go?
It can be difficult to visualize what’s going on deep underground, and this is doubly true where hydrocarbons are involved. The physics get more complicated when multiple liquids are present, and thinking of the subsurface like a big water tank with a few layers holding it underground is a big oversimplification.
This second study brings together our knowledge of how hydrocarbons form in shales and information from the drilling of actual production wells to argue that contaminated water is veryunlikely to migrate out of layers like the Marcellus Shale.
During the “cooking” process that converts organic matter into gas, most of the groundwater in the shale gets driven out. Natural gas is present at very high pressures in these shales, ultimately contained by capillary forces in the tiny open spaces in the rock—the same forces that, for example, enable trees to pull water up into their leaves.
The researchers’ first argument is that there simply isn’t much brine in the Marcellus to seep upward. The fracking fluid that comes back up returns with very high salt content, but the researchers say this can’t be the result of mixing between the fluid and even saltier brine in the shale. If you calculate the concentration of salt the brine would need to have for this to happen, the numbers don’t add up. Instead, the researchers conclude that the fracking fluid is simply picking up salt ions from the rock itself.
And this is where we get to understanding why some of the fracking fluid that goes missing during the fracking process. In our simple “big water tank” picture, we might assume that the fracking fluid mixes into the water already present down there—maybe it even leaks out of the shale. But in experiments with fresh rock samples from a natural gas borehole, they demonstrate the effect they expected to see—the shale soaks up water like a sponge and holds onto it tightly.
This is something that industry professionals know, because the absorption of water can hamper gas production in some settings. However, that hasn’t really been taken into account in most discussions of the risks relating to fracking. Once the fluid soaks into the shale—which it’s able to do partly as a result of lowering the gas pressure—there’s nothing to force it out of the rock to seep upward. After all, there’s a reason the shale has been able to maintain high gas pressure for 200 million years.
It’s not entirely easy to square these results with the studies using salts and noble gases to fingerprint methane in Pennsylvania drinking water wells. The researchers propose an alternate explanation for those salts—that they’re simply picked up as groundwater flows through the drinking water aquifer—but this doesn’t seem to account for all the data. The geological environment is physically and chemically complicated, and it will take more work to make sense of all the details.
Some criticism of this study is inevitable, as one of the three authors is a Shell geologist and some of the funding came from the petroleum industry (in addition to a US Department of Energy grant), but expertise and access to data from the wells that this brings can be very valuable. The industry is often criticized by the public (or frustrated academics) for not providing the data necessary to facilitate research. Studies like this one can help.