Coalbed reservoirs in the U.S. contain an estimated 695 trillion cubic feet (TCF) of natural gas resources, hold 14% (146 TCF) of the total recoverable U.S. natural gas resources, and currently account for 5% (1.0 TCF) of total annual U.S. natural gas production. Accurately determining the initial gas-in-place volume is a crucial element of coalbed gas resource and reservoir evaluation. However, different reservoir property analysis methods often yield very different gas-in-place values. This paper present results from evaluations of the accuracy of commonly used reservoir property analysis methods. Quantitative data for five commonly encountered sources of error will be presented. These errors are: (1) using a too low maximum density limit when using density log data to determine the gross reservoir thickness, (2) using a too low value for the average reservoir rock density, (3) basing the in-situ gas content on gas desorption data collected at ambient surface temperature conditions, (4) basing the in-situ gas content on gas desorption data collected from drill cuttings, and (5) basing the in-situ gas content on the assumption that the residual gas volume value is negligible. These commonly used analysis practices can result in large (50% or greater) underestimation error in the gas-in-place value and, in turn, the cumulative gas recovery predicted by reservoir simulation methods. Because these reservoir property analysis practices have been so widely used in the past, significant potential may exist for large gas resource and recoverable reserve estimate gains in many existing coalbed gas fields.
AAPG Search and Discovery Article #90926©1999 AAPG Eastern Section Meeting, Indianapolis, Indiana