Developing a New Technique for Calculating Accurate Water Saturations from Well Logs in Source Rocks

Abstract

Determining log-based water saturation using Archie's (1942) equation, or any derivative shaly sand method, requires correct inputs to produce valid results. In resource plays, the rock matrix is composed of water wet and oil wet constituents, therefore, correct values of Archie's cementation factor (m) and saturation exponent (n) are critical. In practice, it is pragmatic to use the Pickett plot (Pickett, 1973) to set connate water resistivity (Rw) and Archie's ‘m’. However, it is difficult, if not impossible, to derive Archie's ‘n’ parameter without additional information. Research combining core and log data shows evidence of a positive correlation between Archie's saturation exponent and the total organic content (TOC) in a given unit volume. Using this relationship, Archie's equation may be used to define a variable ‘n’. It is hypothesized that ‘n’ increases with increasing TOC volume as a result of an interruption of electrical pathways that resistivity tools exploit. This disruption results in an increase in the apparent value of ‘n’ required to compute correct water saturations. Due to the apparent excess resistivity in organic-rich rocks, an increase in ‘n’ values or kerogen corrected resistivity is needed to produce a fit to core-derived water saturations. This article will demonstrate the methodology used to derive a variable ‘n’ parameter and kerogen corrected resistivity in an organic-rich interval.