A number of theoretical relationships have been proposed to describe the relationship between resistivity and saturation in shaly sand formations. At the heart of each of these relationships is the segmentation of R0, the resistivity of a water-bearing shaly sand, into at least two terms: the bulk water conductivity and a clay-bound water conductivity which is parameterized in terms of Qv, the cation exchange capacity per unit pore volume. The practical logging implementation of these models has necessarily departed from theory because Qv has been unavailable. Poor estimates of Qv based on "indicator logs' have frequently made the theories seem inadequate. It is now possible to compute an accurate Qv log using geochemical logging data, and from this to derive an R0 curve which can be directly compared with measured resistivity to identify hydrocarbon zones. This new interpretation is tested in two example wells in the Gulf of Mexico and one in a fresh water environment in Kern County, California. The derivation of the Qv log relies on the transformation of elemental concentrations into mineral abundances which are demonstrated to be in good agreement with core data. The mineral abundances are used to derive a matrix density which is combined with a bulk density measurement to produce a significantly improved total porosity log. In the example wells, the porosities are typically 5 porosity units less than the average of neutron and density, and they agree with total porosity measured on cores. The calculated mineral assemblage includes three clay minerals from which it is possible to calculate the cation exchange capacity of the formation; this is also verified with core data. The CEC, matrix density, and porosity logs are used to calculate Qv for input into any of several resistivity equations to derive the RO log. Hydrocarbons are indicated when the measured resistivity is greater than R0. A comparison of the geochemically enhanced R0 logs with those derived using a more traditional perfect shale model without Qv shows a significant improvement in hydrocarbon identification. The traditional approach greatly overestimates bound water saturation and, without adjustment, tends to indicate hydrocarbons through entire intervals. In practice, the perfect shale model would require the log analyst to input physically unacceptable values for connate water and clay-bound water resistivities in order to achieve a satisfactory interpretation. These problems are largely eliminated by using Qv directly.

This content is only available via PDF.
You can access this article if you purchase or spend a download.