Impedance analysis for methane and CO2 discrimination in coalbeds

Diane Jael Lespinasse Fung, Christopher R. Clarkson, Robert James Ferguson

In this paper, we present a methodology to evaluate the use of Elastic Impedance and Elastic Impedance Coefficient to discriminate coals saturated with methane from coals saturated with CO2. We apply this methodology to a dataset of the Fruitland coals Fairway, in the North of the San Juan Basin U.S, which is considered the most productive reservoir of Coalbed methane in the world.

In this project, we develop a proxy model of a 15.25m (50ft) coalbed at a depth of 914.4m (3000ft) over an area of 31.4km2. We obtain the production forecast of 24 wells during primary production and during enhanced coalbed methane by CO2 injection, starting in 1999 and continuing until 2031. Based on the results of the fluid simulation, we perform a Gassmann fluid substitution to model the variation in Vp, Vs and density due to the changes of fluid saturations in the pore space, assuming 100% of brine saturation as the initial state. In this paper we present three cases: the area after several years of depletion, after two years of CO2 injection, and one year after shutting the injector wells. As a result, we observe that the most representative changes are associated with Vp, which presents a decrease of 55-65m/s after shutting the injector wells.

As a final step, we estimate Elastic Impedance and the Elastic Impedance Coefficient. In the case of Elastic Impedance we observe a decrease in the zones around the injector wells that were associated with the movement of the CO2 flood. The Elastic Impedance Coefficient tends to increase around the injector wells, showing the movement of the CO2 front; and it seems to highlight better the effects of CO2 injection. It was not possible to use EI and EC to discriminate CO2 from methane in this study since the values associated to the areas saturated with CO2 coincides with the values associated to some areas saturated with methane. The magnitude of the changes of EI and EC are small and it is difficult to predict whether the changes will be appreciated in seismic data.