Permeability is a key parameter to predict reservoir deliverability and ultimate recovery in tight-gas reservoirs. Since it is a multi-scale property, its values can significantly change with the scale of the medium under investigation. Well-log evaluation and core measurements provide pore-to meter-scale static permeabilities representative of the borehole vicinity, whereas well testing provides kilometer-scale dynamic permeabilities subject to assumptions regarding rock and fluid properties. This paper demonstrates how different ratios between dynamic and static permeability can be used to validate assumptions of reservoir quality and lateral connectivity away from boreholes.
Static permeability (at the borehole) is assessed with extensive sedimentary and petrographic data to quantify the effects of depositional facies, mineral composition, and diagenesis on porosity-permeability relationships. This permeability model is compared to machine learning (ML) permeability using wireline logs to provide additional data in wells where core is unavailable. Absolute permeabilities are converted to gas effective permeability using unsteady state relative permeability testing.
Dynamic permeability is calculated using Pressure Transient Analysis (PTA) from pressure buildup tests in shut-in wells. Over 20 wells with tests achieving radial flow provide the ground-truth for gas effective permeability thickness. Since PTA does not densely cover the field, we use rate transient analysis (RTA) in over 100 wells to derive a pseudo-PTA permeability. This approach includes the estimation of Original Gas In Place (OGIP), which is the most reliable parameter from RTA, using flowing material balance (FMB) and corroborating results by type curve analysis. We finally derive correlations between PTA-derived permeability and RTA-derived OGIP as an additional proxy for dynamic permeability in wells with more than six months of production.
The ratio between dynamic and static permeability provides valuable insights into reservoir architecture and heterogeneity away from boreholes. Ratios below unity reflect the presence of interbedded thin mudstones which provided silica to occlude pore space and throats of nearby sandstones during diagenesis. Mudstones and cemented sandstone shoulders create additional tortuosity that greatly reduces connected gas volumes identified by the transient analysis. Ratios above unity possess fewer mudstone layers and overall better sandstone quality; even though the sandstone thickness can be lower, the dynamic permeability and connected volumes are consistently larger. Lower gross thickness is associated with less accommodation space in the proximity of basement highs, this setting promoted the removal of interbedded mudstones (and associated damaging effects on reservoirs) and boosted reservoir lateral connectivity. The understanding of dynamic and static permeability ratios; and their linkage to diagenetic effects on depositional architecture contribute to identifying undeveloped resources (i.e., infill drilling) as well as better prediction of initial reservoir performance in new wells.
