Fracking, Natural Gas in Groundwater, and Stable Isotope Forensics

In Part 1 of my post on Hydraulic Fracking and Natural Gas in Groundwater, I explained how the composition of natural gas establishes a basis for differentiating between thermogenic and biogenic gases. Recall that thermogenic gases are formed from the heating of organic matter in high-temperature subsurface environments. Gases that form under these conditions are typically composed of hydrocarbons ranging from one to six carbon atoms, or methane (C1) to hexane (C6).

Biogenic gases form in lower temperature regimes, usually in the near subsurface, as a result of microbial activity on buried organic matter. These gases, which are dominantly methane with very minor ethane (C2) and propane (C3), are commonly associated with coalbeds, swamps, and landfills.

We can differentiate between thermogenic and biogenic gases by considering the molar ratio of methane to ethane and other hydrocarbons (C1/(C2+C3). As noted in my previous article, it is not uncommon for that ratio to be less than 100 among thermogenic gases and to be greater (oftentimes, much greater) than 1,000 among biogenic gases. The difference in ratios is a function of the deficiency of ethane (C2), propane (C3), and higher-chain hydrocarbons in biogenic gases.

m1_768
Figure 1

 

In addition to the molar ratios, geochemistry offers a deeper look into the signatures of natural gases by mass spectrometry.  This gives geochemists a powerful means of differentiating between thermogenic and biogenic gases. Specialized laboratories have the equipment needed to conduct such analyses on samples of dissolved gas in water and samples of gas collected from production wells. Such analyses focus on specific carbon and hydrogen components of the methane (C1) fraction.

More specifically, laboratories measure the abundances of an isotope of carbon (carbon-13 or 13C) and an isotope of hydrogen (deuterium, 2H or D) in methane, relative to their abundances in standards determined by the International Atomic Energy Agency. For 13C, the standard is known as Vienna Pee Dee Belemnite (VPDB) and for D, the standard is Vienna Standard Mean Ocean Water (VSMOW).

In a substance such as natural gas, methane (C1) will be depleted in both 13C and D, with respect to VPDB and VSMOW, and the abundances of the two isotopes in methane are reported as ratios in units of parts per thousand, or per mil (e.g., δ13C‰ VPDB and δD‰ VSMOW). Because natural gas is depleted in 13C and D compared with the VPDB and VSMOW standards, the reported abundances will be negative.

Figure 1 is a plot of the δ13C and δD values for the two samples of Wilcox dissolved gas and the five samples of Haynesville formation gas as discussed in the previous post. There are three shaded fields: One is representative of thermogenic gas and the two other fields are representative of gas produced by microbial activity. The microbial gas field to the left of the thermogenic gas field is the domain within which the δ13C and δD values of gas associated with lignite and other low-rank coals cluster. The shaded area below the thermogenic gas field is representative of microbial gases that form in swamps and landfills.

m2_768b
Figure 2

All of the Haynesville gas samples lie near the center of the thermogenic gas field and the two samples of dissolved gas from the Wilcox aquifer are within the field associated with coalbeds. This separation between the Haynesville and Wilcox gas samples is unmistakable and establishes a second and very sound geochemical basis for differentiating between thermogenic and biogenic gas. Geochemists look for separations such as this when trying to identify the conditions under which a gas originated.

We can take this a step further by plotting δ13C against C1/(C2+C3), as shown by Figure 2 (also known as a Bernard plot). In this figure, the Haynesville samples lie within the thermogenic field, in the lower right. The C1/(C2+C3) ratios are all less than 100 and the associated δ13C values range from approximately –45‰ to –40‰. The two Wilcox gas samples are located within an area of the Bernard plot dominated by microbial activity. For these samples, C1/(C2+C3) is close to 10,000 and δ13C is –68‰ to –64‰. The separation between the Haynesville and Wilcox gases is so distinct on this plot, both with respect to δ13C and C1/(C2+C3), that there can be no reasonable basis for arguing that there is any influence of thermogenic gas in shallow groundwater.

In this brief presentation and the previous article, we have examined some very basic concepts that offer penetrating insight into the origin of natural gas discharging from shallow ground­water. The concepts and associated plotting methods that I have illustrated are well established and widely used. It is especially important to consider the overall composition of gas and the isotopic signatures of gas discharging from domestic wells before concluding that oil and gas operations are to blame.

 

Share and Enjoy

  • Facebook
  • Twitter
  • Delicious
  • LinkedIn
  • StumbleUpon
  • Add to favorites
  • Email
  • RSS

No Comments

Post a Comment