Increasing Coal Absolute Permeability in the San Juan Basin Fruitland Formation

Summary Recently measured data show that the absolute permeability of coal natural fracture is increasing significantly with continued gas production in the San Juan basin Fruitland formation. This phenomenon caused gas-production rates to be many times greater than expected from early production history. The phenomenon also caused producing bottomhole pressures to increase when gas rates were constant, opposite from that expected from conventional applications of Darcy's law. The increase in absolute permeability caused by gas desorption has been measured on cores,1–3 but, until recently,4 there was no verification that this phenomenon occurs in situ. Palmer and Mansoori4 (P&M) presented a new theory and showed how this theory matched gas- and water-production rates and estimated bottom-hole-pressure data for a high-deliverability San Juan basin Fruitland formation coal-gas well. However, Palmer and Mansoori4 had no transient pressure data to support in-situ permeability changes. This paper documents data from drillstem tests (DST's) and shut-in tests with analyses thereof and additional production-rate and pressure behaviors that support the P&M theory. The well-test data were measured in three wells completed in the San Juan basin Fruitland formation coal seams located under the Valencia Canyon (VC) area. These wells, VC 29-4, VC 32-1, and VC 32-4, are located in Sections 29 and 32, T33N, R11W, La Plata County, Colorado, and operated by EnerVest San Juan Operating LLC. In addition to the well-test data, EnerVest and the Gas Research Inst. (GRI) collected extensive formation-evaluation data and performed detailed analyses that allowed a thorough description of the area. Although there are other wells operated by EnerVest in the area, well-test data were not available from the other wells to determine the absolute permeability estimates; therefore, these wells are not discussed in this paper. The P&M theory was calibrated with well-test-derived absolute permeability estimates and published coal-shrinkage data.5 Reservoir simulation models, based on the calibrated theory, matched the unusual producing, bottomhole-pressure behavior. Without the P&M theory it was not possible to match pressure behavior or permeability estimates with reasonable variations of reservoirproperties input into the reservoir-simulation models. The remainder of this paper summarizes the well-test analysis results from the three wells and, for brevity, one set of well-test data and one simulation study. The well-test data and simulation studies for the other wells were similar to the examples.