Northeastern Section - 50th Annual Meeting (23–25 March 2015)

Paper No. 2
Presentation Time: 1:30 PM-5:30 PM

THE MICROBIAL SLICE OF THE NATURAL GAS SUPPLY: CAN THIS CONTRIBUTION BE INFERRED FROM ANALYSIS OF DECLINE CURVES IN THE ANTRIM SHALE, MICHIGAN BASIN?


FERSTAD, Johannes O., Amherst College, Amherst, MA 01002, MARTINI, Anna M., Geology Department, Amherst College, 11 Barrett Hill Road, Amherst, MA 01002 and HORTON, Nicholas J., Department of Mathematics and Statistics, Amherst College, Amherst, MA 01002, jferstad15@amherst.edu

Over the past few decades, the importance of microbial contributions to our natural gas supply has been widely recognized. Conditions for microbial methanogenesis are fairly well constrained: temperatures up to ~80oC, low sulfate concentration, and chloride concentrations under ~2M. However, these are rather expansive boundaries and within each range there is constant turnover in population density as well as specific microbial abundances. In addition, the rates at which these microbes are able to convert complex organic matter into methane also depend upon these environmental conditions.

The Antrim Shale, a Devonian black shale, has been producing both microbial and thermogenic gas for many decades. One line of evidence for microbial gas generation is the correlation of the deuterium in the water to that in the methane, following a dominantly CO2-reduction methanogenic pathway. The gas composition has changed in lockstep with the water chemistry over ~25 years of production. This may signify a relatively high proportion of recently produced methane in the system.

Another potential way to gauge the importance of the biogenic contribution to Antrim Shale gas production is by examining decline curves that cross the strong geochemical gradients that exist. For example, Cl concentrations go from potable to nearly 6M, and ethane (a thermogenic gas component) goes from undetected to ~15 % of the total gas volume produced. Decline curves may suggest that areas with more biogenic gas generated have, as a result, flatter slopes.

Initial analysis of Otsego County decline curves identifies a weak but significant trend. Breakpoint models were used to estimate decline curves for each production unit, including their boundary-dominated flow decline rates. Weighted multiple regression models find that production declines more steeply for wells in northern parts of the county. The trend is robust to exclusion of wells with questionable fits and to controlling for a number of relevant factors. But this trend might capture reservoir properties that develop due to differences in the depth of production. At the shallower northern end of the play pressures are lower and the fractures are more open leading to potential gas losses and less initial storage. Controlling for these and other variables is ongoing.