Quantitative Evaluation of Coal Seam Gas Content Estimate Accuracy

Abstract

Abstract The proper measurement and analysis of gas desorption data from freshly cut coal samples is crucial to reliably predict the gas production performance and reserve potential of coal seam gas reservoirs. The technical literature offers may procedures for evaluating the total sorbed gas content of coal samples. This paper presents evaluations of the accuracy and comparability of gas content estimates for freshly cut San Juan Basin Fruitland Formation and Piceance Basin Cameo coal samples obtained using commonly practiced analysis procedures. Benchmark (or "ground truth") gas content values were assembled from pressure core and sorption isotherm data for comparison against gas content estimates obtained from conventional whole core, sidewall core, and drill cutting samples. Lost gas content estimate accuracy determined by the Direct (U.S. Bureau of Mines) Method,1,2,3,4 the Smith & Williams Method,5,6 and the Amoco Method7,8 was evaluated and compared. Assessments were also made of the sources and magnitudes of sampling and procedural method errors on the accuracy of the gas content estimates. The choices of sample type, gas desorption temperature conditions, and lost gas analysis method significantly affected the accuracy of the gas content values. The Direct Method analysis of conventional whole core, reservoir temperature gas desorption data provided the most accurate gas content values. Using ambient temperature gas desorption conditions rather than reservoir temperature conditions is the greatest source of error in gas content estimates.