A Hydraulic (Flow) Unit Based Model for the Determination of Petrophysical Properties from NMR Relaxation Measurements

Abstract

Abstract This paper presents preliminary results of a Hydraulic (flow) Units based research program aimed at defining more effective guidelines for the use of nuclear magnetic resonance tools (NMRT) for petrophysical evaluation. Successful application of the Hydraulic Units concept for predicting petrophysical properties from wireline logs has been demonstrated previously. In this paper, we show obvious but crucial relationships between the Hydraulic Units concept and nuclear magnetic resonance (NMR) relaxation data. The linkage between these two technologies is possible because the hydraulic unit zonation indicators referred to as flow zone indicators (FZI) and NMR relaxation data are related to the surface phenomenon which controls the microscopic attributes of the rock. It is shown that although NMR relaxation measurements respond to pore body size rather than pore throat size, an effective NMR based permeability model, which includes pore size to pore throat size ratios, can be developed. The phenomenological model presented herein consists of mathematical relationships between FZI and NMR relaxation data, specific surface area, tortuosity and relaxivity (ρ). Laboratory (NMR) relaxation time measurements on CORESPEC-1000 together with core analysis data of permeability, porosity, centrifuge capillary pressure and thin section petrography have been used to validate the model. The results show that the observed variability in the relaxivity constant (ρ) is due to the existence of different hydraulic units in the reservoir rock. We also demonstrate that relaxation time cut-off (T1c) used for determining producible fluid from relaxation time distributions may vary slightly as a function of rock types. Hydraulic units based predictive algorithms for the estimation of relaxivity (ρ) and T1c, using pore throat size distributions (mercury injection data), and centrifuge capillary pressure data are validated. A more effective permeability estimation algorithm is also presented.