An Integrated Approach for Reservoir Modelling of Complex Miocene Fan Delta Sandstone, El Morgan Giant Field: A Case Study from Gulf of Suez

Stock Tank Oil Initially in Place (STOIIP) uncertainty has lately been mentioned as a key influence on mature oil field development plans under the present difficult economic conditions. In the Gulf of Suez, conformance management of multilayered Belayim and Kareem reservoirs with substantial lateral and vertical variability is extremely challenging. In such heterogeneous reservoirs, reservoir productivity and injectivity require a high-quality 3D geo-cellular model. El Morgan oil field embraces an area of about 46 sq. kms. The field attained its development maturity. So, this was the driver to look intelligently for new approaches to maximize the production and additional reserves with least cost. There are two separate hydrocarbon columns in the Morgan field, the Belayim and Kareem formations, respectively were deposited during active Gulf of Suez rifting. This work focuses on the Belayim reservoir, which is a mature deltaic clastic deposits with variable heterogeneously distributed argillaceous matter. Sedimentologically, five facies of Belayim Formation were recognized and comprise a complex series of deltaic environments ranging from upper fan delta to distal. The sandstones dominantly range from very fine to very coarse-grained and are only weakly to moderately overprint by authigenic cements. Moreover, a machine learning approach was applied to predict facies and its reliability versus cored wells, and showed up to 73% accuracy. The integrated evaluation of Belayim reservoir showed that the sandstones is essentially controlled by the primary depositional detrital clay content, with additional control by grain size. The variations in pore system characteristics have been used to define "rock types," which show a good correlation with the reservoir quality trends from the core descriptive lithotype scheme, and can potentially be integrated with MICP data. These rock types show distinct ranges of occurrence within the individual facies and hence allow reasonable predictability of likely reservoir character within the sedimentological framework. The resultant layering scheme was integrated with all available dynamic data to best define the reservoir architecture, and the flow units to construct a fine-grid 3D geological model. An Object modelling method is applied to distribute facies in the model. Porosity modeling is conducted and is subsequently used as a guide for permeability modelling. The model results show that the reservoir quality controls optimized the water flooding, particularly in the north block, which is yet to meet the production expectations compared to the south block. Therefore, we recommend the re-completion of the M1-118 well from Kareem injector to be a Belayim producer, with an estimated initial rate of 500 BOPD. We further recommend replacing the scaled up peripheral injector well M1-207ST1 (not accepting water) with a side-track at a better location based on the facies model for restoring the water injection in the northern polygon of north Morgan Belayim. This will improve the sweeping efficiency with estimated adding reserve about 0.8 MMSTBO.