Calibrating NMR Response to Capillary Pressure Curves in Fine Grained Lithologies, Pretty Hill Formation, Otway Basin

ALI ALGHAMDI, M.Sc Candidate,  Australian School of Petroleum, University of Adelaide, South Australia,  [email protected]

Nuclear magnetic resonance (NMR) tools are commonly used in formation evaluation. NMR T2 distribution data have been used by previous authors to build down-hole pseudo capillary pressure curves in reservoir quality rocks. The objective of this study is to generate NMR-derived down-hole pseudo capillary pressure curves in very fine grained lithologies in order to determine whether it is possible to estimate capillary displacement pressures and thereby sealing capacity.

NMR T2 relaxation time distributions of flood plain facies at Redman-1 were converted to pseudo capillary pressure curves. The generated curves were compared with mercury injection capillary pressure (MICP) curves. Displacement pressures were selected from both pseudo capillary pressure curves and actual MICP curves at different percentile cutoffs of non-wetting saturations. The best cutoff in displacement pressure estimation is the 20^th percentile with correlation coefficient of 0.59. Statistically, the correlation coefficient of the 20^th percentile is too low for calibration.

The reason for the lack of robust calibration is related to the actual properties of the rock: the Redman-1 flood plain samples have high iron contents (Fe2O3 content ranges between 5.2-7.16 wt%) with increased magnetic susceptibility and elevated internal field gradients. The NMR T2 response is affected significantly by the internal magnetic field gradient which depends on the magnitude of magnetic susceptibility. Surface relaxivity changes and high pore to throat size ratio also contribute to the difference between the two measurements. 

The main conclusion is that using NMR response to estimate displacement pressures in iron-rich very fine grained rock is not a viable methodology.