Improving the Permeability Derivation from NMR Data for Reservoir Rocks with Complicated Pore Connectivity

Abstract

The formation permeability is one of the most important Petrophysical information acquired by low field Nuclear Magnetic Resonance (NMR) data. The absolute permeability of reservoir rocks can be accurately calculated by porosity measured by NMR if permeability is well correlated with porosity [1, 2]. This is true for the rocks with all pores are well connected, such as conventional sandstone reservoir. The correlation between permeability and porosity, however, does not always established well for carbonate reservoir rocks, which are known to have heterogeneous pore-to-pore connectivity. Thus, without accurate pore connectivity information, the permeability derived from NMR data for rocks with poor connection among pores is not quite accurate. The current study proposes a new method to measure an accurate pore-to-pore connectivity from the total volume of a fluid displacement through miscible single phase fluid mixing process. The pore connectivity factor is calculated from the fluid connectivity through different pore systems which can be derived by the volume comparison of NMR invisible D2O replacement with NMR visible H2O in each pore systems. Once the pore connectivity is measured, it is applied to the modified permeability model, such as modified free-fluid model (Timur-Coates model) [3], to calculate accurate permeability. Another benefit of the proposed method is the flexibility of D2O injection condition. Carbonate reservoir rocks with various degrees of connectivity have been studied in the current study. For the verification purpose, the permeabilities calculated by the proposed method have been compared to those measured by other conventional laboratory techniques. The proposed method can also be applied to NMR logging technique for accurate NMR driven permeability derivation which currently accepted as references for reservoirs with complicated connectivity, such as carbonates.