This paper presents a case study of core-to-log data integration of a well drilled in a carbonate formation bearing bimodal pore system. Based on the evidences, a new approach is proposed for the interpretation of Nuclear Magnetic Resonance (NMR) logging data. The main tasks involved in the data integration process were:
interpretation of available basic and special core analysis data (porosity/permeability of 163 plugs, Mercury Injection Capillary Pressure (MICP) data of 43 samples and wettability tests),
NMR measurements in 9 plugs at different saturations (100% brine and Swi) for three different wettability conditions (plugs were aged in oil for 18, 33 and 42 days) and,
integration of the lab results with NMR well logging data.
The main objective of the study was to understand fluid distribution in the bimodal pore system, in order to reduce the uncertainties associated with water saturation and permeability determinations from the NMR log interpretations.
The first NMR log interpretations of the well using conventional cutoff (~ 100 ms) resulted in high water saturations. Such saturations were incompatible with the laboratory results. Core analysis from the key cored interval have shown the rocks to be of grainstone texture and composed of rounded grains with high porosities and permeabilities. The analyses also point to a high volume of micropores which were confirmed by the MICP analysis and by the transverse relaxation time distribution (T2) in the 100% brine saturated plugs. The macro and microporosity proportions determined by both methods (NMR and MICP) confirmed each other, indicating proportionality between the amount of pores and throats. In addition, macroporosity revealed a good correlation with absolute permeability, allowing to conclude that macroporosity is the dominant parameter controlling fluid flow. For the analysis of T2 distribution at irreducible water saturation (Swi) and different aging times, rock samples were displaced with a high viscosity oil (110 cp at 90°C) whose transverse relaxation time (T2) was lower than the two distinct micro and macropores modes. Results revealed that with aging macropores and part of the micropores were sequentially filled by oil.
This large amount of microporosity filled by oil affects the T2 distribution and therefore the determination of cutoffs (T2 cutoff), since these micropores, related to Bulk Volume of Irreducible Water (BVI), may contain oil but do not have significant contribution to permeability.
This study shows that, depending on the formation wettability and porosity distribution, the use of a second cutoff in the NMR interpretation solves the incompatibility problem between permeability and porosity of the wells obtained by NMR logs.
