Integrating NMR With Other Petrophysical Information to Characterize a Turbidite Reservoir

Abstract

Abstract Uncertainty lies in developing the nation's discovered oil and gas fields. Since the subsurface cannot be accessed directly, indirect methods and procedures must be developed to estimate the size and productive potential of a reservoir. The value of individual pieces of information is realized when they are integrated with other sources of information to characterize a reservoir. New and improved technologies e.g., Nuclear Magnetic Resonance1 (NMR), has been extensively used to determine the reservoir properties such as lithology independent porosity, permeability, water saturation, irreducible water saturation etc. The Free Fluid Model2 and the Mean T2 Model3 are two NMR based methods to estimate permeability. Partitioning of pore fluid into free and bound volume depends on the precise value of transverse relaxation time (T2), T2cutoff. Infact, for a water wet rock, bound volume irreducible (BVI) is the amount of water that coats the surface of the pore space and can be referred as the irreducible water saturation (Swirr). In this study fifteen plug samples from a turbidite reservoir (Grubb formation) from California were used. The plugs are classified as arkosic-lithic sandstones (with an average mineral composition of feldspars (52%), quartz (40%), clays (4%), carbonates (3%) and others (1%)). The porosity ranges from 6% to 19% with an average of 15% and the permeability ranges from 0.045 md to 30.42 md having an average of 9.7 md. NMR spectra were collected on saturated (25,000 ppm NaCl brine) and desaturated samples. Using the cutoff BVI (CBVI4) technique for estimating the percentage of bound water, the measured T2cutoff for the fifteen plugs ranged from 5 msec to 15 msec. The computed free fluid index (FFI) based on the above T2cutoff range is approximately 30% to 50% higher than the FFI calculated from industry recommended T2cutoff value of 33 msec for clastic rocks. For this reservoir, the Mean T2 Model3 gives better estimate of permeability as compared to those calculated using the Free Fluid model2. The irreducible water saturation (Swirr) decreases with increase in permeability. The comparison of Swirr from capillary pressure measurement and centrifuge experiments showed a pattern that can be explained on the basis of mineralogy and microporosity. The main purpose of establishing a correlation between NMR T2 distributions with mercury injection capillary pressure (MICP) is to generate the capillary pressure curves directly from NMR relaxation time distribution during wireline logging. An average value of effective surface relaxivity (ρe) of 24.3 μm/sec has been established for the Grubb formation, which is then used to generate the pseudo capillary pressure curves. The NMR derived capillary pressure curve shows a reasonable match with the air-brine capillary pressure curves derived from MICP.