Marie Van Steene, SPE, and Mario Ardila, SPE, Schlumberger; Richard Nelson, SPE, and Amr Fekry, SPE, BP Egypt; and Adel Farghaly, SPE, RWE Dea Summary In hydrocarbon reservoirs, fluid types can often vary from dry gas to volatile oil in the same column. Because of varying and unknown invasion patterns and inexact clay-volume estimations, fluid-types differentiation on the basis of conventional logs is not always conclusive. A case study is presented by use of advanced nuclear-magnetic-resonance (NMR) techniques in conjunction with advanced downhole-fluid-analysis (DFA) measurements and focused sampling from wireline formation testers (WFTs) to accurately assess the hydrocarbon-type variations. The saturation-profiling data from an NMR diffusion-based tool provides fluid-typing information in a continuous depth log. This approach can be limited by invasion. On the other hand, formation testers allow taking in-situ measurements of the virgin fluids beyond the invaded zone, but at discrete depths only. Thus, the two measurements ideally complement each other. In this case study, NMR saturation profiling was acquired over a series of channelized reservoirs. There is a transition from a water zone to an oil zone, and then to a rich-gas reservoir, indicated by both the DFA and the NMR measurements. Above the rich gas, is a dry-gas interval that is conclusively in a separate compartment. Diffusion-based NMR identifies the fluid type in a series of thin reservoirs above this main section, in which no samples were taken. NMR and DFA both detect compositional gradients, invisible to conventional logs. The work presented in this paper demonstrates how the integration of measurements from various tools can lead to a better understanding of fluid types and distribution.
Computing clay volume using elemental neutron capture spectroscopy logs in combination with a multimineral solver for the complex, shaly sand reservoirs of the Nile Delta reservoir improved accuracy over using the mineral fractions output from the spectroscopy model alone. It was also found that the aluminium log from direct aluminium yield measurement leads to a better clay volume estimation, as opposed to using the aluminium log from the aluminium emulator algorithm.
Combining the spectroscopy data with borehole image data generated a high-resolution lithofacies column that provides an accurate stratigraphic interpretation. Applying cutoffs to generate a high-resolution sand count enabled us to sort the reservoir units from the poorest to the best quality sands and improved our understanding of the distribution of the best reservoir quality in the well. This approach provides a unique solution to characterize thinly bedded reservoirs in wells drilled with oil-based mud.