Accurate Reservoir Water Saturations from Oil-Mud Cores: Questions and Answers from Prudhoe Bay and Beyond
Richard Woodhouse: Petrophysical Consultant

Abstract: Direct reservoir formation water extractions on more than 8,000 plugs from 27 wells cored with oil-base mud were used to determine water saturation (Sw) in Prudhoe Bay field, Alaska. A question and answer format is used to review the abundant and compelling evidence from the multicompany research program that established their validity. The main criticism of oil-mud core Sw is that water might be lost while the core is cut and brought to the surface. This criticism is easily asserted but difficult to disprove. New technology coring bits made possible the recovery of cores that have no invasion at their centers. Photographs clearly show these uninvaded centers. Invaded and uninvaded cores were found to have the same Sw values. From pressure-retained cores, it was determined that depressurization did not cause any measurable water loss. Like most reservoirs, Prudhoe Bay has a very long transition zone and, in all but the lowest section, the evidence demonstrates that the core Sw values are valid. Sw values at Prudhoe Bay vary from 1% to more than 50%. The oil-mud core Sw values were used to calibrate Sw values from the resistivity logs and the Archie equation by adjusting the saturation exponent (n) value. This calculated n varied from 1.7 to 3.1. Three indirect Sw evaluation methods (log, capillary pressure, and chemical tracer-test analyses) give results consistent with the core Sw values after painstaking work to understand their many variables. For example, the reservoir water salinity varies from 6 to 45+ g/L of Cl, and Rw varies from 0.13 to 0.72 ohm-m at 68°F, when the aquifer is 0.34 ohm-m. Standard laboratory measurements of n have large uncertainties because of the difficulty of reproducing both the reservoir water saturation and its distribution. The uncertainties of Sw from oil-mud cores and the Archie equation are evaluated by partial differential analysis. The practical importance of standard log interpretation is not diminished by this work. This paper, and the teams other publications, show that a significantly more accurate Sw measurement is available when the extra effort is justified. Subsequent work elsewhere with oil-emulsion mud has provided cores of equal validity to those in Prudhoe Bay field. Every time hydrocarbon-zone cores are cut in oil-base muds of any type it is clearly worthwhile to obtain good quality core Sw measurements.

Dielectric Dispersion of Partially Saturated Porous Media in the Frequency Range 10 Hz to 10 MHz
Ali A. Garrouch: Department of Petroleum Engineering, Kuwait University Mukul M. Sharma: Center for Petroleum and Geosystems Engineering, The University of Texas at Austin

Abstract: Two- and four-electrode setups have been developed to measure the complex impedance of partially saturated Berea sandstone, tight-gas-sand rocks, and Ottawa sand-bentonite mixtures saturated with n-decane and NaCl brine solutions. The effects of water saturation, clay content, and frequency on both the real and imaginary parts of the rock impedance are investigated between 10 Hz and 10 MHz. A reactivity index exponent (analogous to the resistivity index exponent) is defined and shown to be related to the water saturation by an Archie-type relation. A linear relationship between the resistivity index exponent and the reactivity index exponent has been observed experimentally at frequencies below 0.1 MHz and has been deduced algebraically as well. The reactivity index exponent can be used to estimate water saturations at frequencies above 0.1 MHz where the resistivity index exponent is found to be inadequate. A generalized Maxwell-Wagner model, which accounts for double-layer dielectric dispersion, is used to explain the experimentally observed trends. As the clay fraction is increased by a factor of 2, the dielectric constant of a simulated sand pack is found to increase by approximately the same factor at almost all frequencies from 10 Hz to 10 MHz. This linear dependence between rock dielectric constant and clay percentage has also been observed with the two-electrode data for tight-gas-sand samples of similar porosities and permeabilities.

An Explicit Form of the Waxman-Smits Equation for Shaly Sands
Peter U. Ohirhian: University of Benin, Nigeria

Abstract: An explicit equation that can calculate water saturations for shaly sands within ± 0.3% error has been developed and is compared with water saturations from the Waxman-Smits (WS) equation. The formula developed here is simple and can easily be applied by log analysts not familiar with numerical analysis. Furthermore, evaluation of the new equation requires a shorter time than the iterative methods commonly used for solving the WS equation. The accuracy of the new equation has been demonstrated over a wide range of log data (Freedman and Ausburn,1985) representative of shallow shaly sands of the Gulf Coast and shallow freshwater California shaly sands.

Using NMR Logs to Reconstruct SP and Confirm Rw
G. R. Coates and D. Miller: Halliburton Energy Services, Houston, Texas, USA

Abstract: New nuclear magnetic resonance (NMR) measurementssuch as mineralogy-free porosity, direct hydrocarbon volume and type, and clay-bound water and its tie to cation exchange capacity (CEC)have brought a new level of quality to log-derived water saturation, both in the flushed zone and in virgin formation. However, the inversion of resistivity logs into water-filled porosity remains heavily dependent on formation water resistivity (Rw). On many occasions, the log analyst has no direct measure of Rw, and the Rw estimate that the analyst must make can introduce a high degree of uncertainty in subsequent calculations. Such a situation led to a search for a method that could be used to help ensure high-quality Rw selections. A method was found that integrated NMR and spontaneous potential (SP) measurements.

Identification of Gas-Oil Contacts by Overlay Techniques Using Wireline Log Data
Birbal Singh, R. V. Rao, and P. P. Deo: Oil and Natural Gas Corporation, SRBC, Madras, India

Abstract: The Tiruvarur structure is in the southwestern part of Karaikal High in the Cauvery basin (Onland) of southern India. Seven wells have been drilled on this structure to date, five of which are oil or gas bearing. Initial production test results indicate the pay zones have gas-cap gas, oil, and water. However, these fluid interfaces are not easily discernible on well logs. Sedimentological and mineralogy studies on cores and cutting samples in these wells indicated minerals such as feldspar, biotite mica flakes, and ferromagnesians. These minerals appear to have affected the responses of density, neutron, and gamma ray logs, thus making it difficult to identify gas-oil contacts (GOC) and oil-water contacts (OWC) with a cursory look at the well logs. A methodology has been adopted to identify the fluid contacts through the study of well logs and various lithology crossplots, F overlays, and DT-PHIN overlays. The crossplots clearly indicate the influence of heavy minerals in suppressing the gas effect and the overlay techniques help demarcate precise fluid contacts. The results agree closely with the testing results. Using this methodology will help in the formulation of an optimum perforation policy that avoids gas or potential gas-oil ratio (GOR) zones, thereby resulting in improved oil recovery.