In general, producing gas-oil ratio (GOR), mobility, and reservoir pressure trends observed during the depletion of conventional oil reservoirs produced above and below bubblepoint pressure are quite well understood. However, for tight/shale (unconventional) light oil reservoirs developed using multi-fracture horizontal wells (MFHWs), the case is different — the GOR may exhibit a greater variety of behaviors caused by: complex sequences of flow regimes (transient, transitional, and boundary-dominated flow), variable operating conditions, reservoir/fracture heterogeneity, and fluid properties, amongst other factors. As a result, the relationship between GOR and well performance is less well understood for unconventional reservoir depletion scenarios. The objective of this work is to study the various factors which control GOR trends, and then to relate these factors to well performance using field examples and realistic reservoir simulation cases.
Semi-synthetic simulation cases generated using field-derived flowing bottomhole pressures, history-matched simulation results, and field cases are provided in this work to evaluate the relationship between GOR and well performance. The semi-synthetic cases are generated for a range of fracture/reservoir geometries — including a variable-fracture spacing (multi-fracture) scenario in order to observe the effects of fracture spacing differences. Matches for each of the cases to dimensionless type curves provide a diagnostic to visualize the differences in well performance for each fracture/reservoir geometry scenario and for single-phase versus two-phase flow. Additional reservoir simulation cases are provided to (1) demonstrate that the effects of multiphase flow can be corrected for using dimensionless type-curve matching and (2) interwell communication can affect GOR trends.
A suite of well performance data cases from the Avalon and 1st Bone Spring formations in the Permian Basin (TX/NM, USA), which illustrate a wide range in GOR profiles, is studied. Comparisons of the well performance and GOR data with the semi-synthetic and reservoir simulation results provide conceptual and illustrative guidance for the comparative analysis of such data. Additional field cases from the Wolfcamp shale (TX/NM, USA), Montney siltstone (BC/AB, CA) and Eagle Ford shale (TX, USA) reservoirs are also provided as examples.
The semi-synthetic cases studied exhibit an adequate range in GOR profiles which are affected by the fracture/reservoir geometry and these cases can be related to differences in flow-regime sequences and timing. The magnitude of deviation of well performance during two-phase flow from the single-phase flow case, as determined by comparison of these cases matched to dimensionless type curves, also depends on flow-regime sequence — the greatest differences are observed when boundary-dominated flow occurs.
The Avalon and 1st Bone Spring wells, binned according to GOR level (high, medium, low), exhibit a range in well performance (decline) characteristics, and demonstrate that the deviation of well performance from single-phase dimensionless type curves is qualitatively related to its GOR bin (or category). However; EUR, as determined from decline-curve analysis, is not directly related to GOR — and moreover, the EUR differences observed in this work are small given the wide-range in GOR profiles.
Given the results of cases studied in this work, we provide recommend that "rules of thumb" and simplified conclusions regarding the relationship between fluid ratios and well performance be avoided — such as inferring the future decline rates of oil production based upon GOR, as such conclusions are not clearly proven from our modeling work, nor our observations of well performance behavior.