Evaluating the Impact of Porosity–Permeability Heterogeneity on Oil Production in Carbonate Reservoirs

Abstract

Over 60 % of the world's conventional oil is found in carbonate reservoirs; however, their highly complex pore structures have created some of the poorest correlations between porosity and permeability. Because there is great ambiguity regarding how each pore type contributes to production rates (i.e., the amount of fluid produced from the reservoir), conventional models are unable to account for the heterogeneity of carbonate reservoirs, which can lead to the placement of inappropriate wells and miscalculations in production forecasts. This research will quantify how variability in porosity  and permeability  affect oil production rates through the use of integrated core analysis, wellbore data analysis, and dynamic reservoir simulation. Three carbonate fields were used to collect the data for this project: one from the Middle East, one from North America, and one from Europe—for a total of 500 core plugs sampled for helium porosity and Klinkenberg corrected permeability (between 1970 and 1999). Additional data will also be collected from thin-section petrography, micro-tomography (CT scanning), and mercury injection capillary pressure analysis. Pore type classifications were achieved and numerical reservoir simulations (CMG IMEX) and statistical regression (including power or Kozeny Carman) were completed under different scenarios . The analysis has found that there are many types of porosity (vugs & fractures) in secondary porosity that dominate production while decreasing the predictability of ; and the introduction of new empirical coefficients relating to vug connectivity to effective permeability improves production forecasting. In terms of production rate sensitivity, at higher levels of porosity, it appears that the effect of permeability is three times greater than that of porosity for carbonate rocks with permeabilities greater than 10 mD. In light of these findings, I propose that a pore type-based correction factor to Darcy’s equation be developed, as well as provide completion strategy recommendations for the carbonate formation. The resulting multivariate regression (R² = 0.81, MAPE = 18 %) is superior to the performance of existing models and is considered to be a clear, physically based function for use by reservoir engineering staff.