Water Quality Profiles From Well Logs: When They Don't Match Lab Analyses, Consider the Conductive Clays

Abstract

In recent years, documenting water quality in shallow aquifers has become a critical issue for oil and gas operations. Laboratory analyses on water samples are necessary for water quality work, but they are expensive and often difficult to obtain where they are needed. Well logs are available in most areas, and the use of formation water salinity/resisitivity data is routine since it is required for obtaining saturation data from log analysis. Water resistivity can easily be related to salinity as NaCl, and TDS for more complicated chemical profiles can be estimated with information on the ion species likely to be present in the water. Over the years, a number of simple methods have been proposed for estimating water resistivity (Rw) from logs, which are adequate for the high-TDS and dominantly NaCl water found in many oilfields. However, these methods can lead to inconsistent or misleading results when applied to low-TDS aquifers with waters whose chemistry is not primarily NaCl. Often the reason for this is the presence of clays which contribute an increasing fraction of the formation conductivity as water salinity declines. Extensive research on cores revealed a temperature- and salinity-dependent relationship between the cation exchange capacity of the clays and the salinity of the equilibrating brine solution (formation water). Petrophysical methods based on this research have been applied to derivation of hydrocarbon saturation in O&G reservoirs having a wide range of salinities and temperatures with great success. With proper methodology, the same principles can be applied to determination of water quality in fresh to brackish aquifers. Where only water is present, water saturation models simplify since Sw= 1.0 and the saturation exponent is not needed. The remaining variables are the contributions to formation conductivity from water and clay. Of the well-known shaly-sand log analysis equations, the Waxman-Smits model is most applicable to fresh-water systems since the non-linear behavior of the clay conductivity in low-TDS brines is best documented for this model in published research. By properly accounting for the contribution of the clays to formation conductivity, the accuracy of water quality profiles from logs can be significantly improved over those derived from SP logs or simple Archie-based calculations. The log-based water quality profile can be further refined by accounting for the varying electrical activity of the non-NaCl ions usually present in low-TDS waters. The analytical model for water quality should be calibrated to laboratory water analyses, in concept analogous to calibrating reservoir petrophysical models to core data.

AAPG Datapages/Search and Discovery Article #90266 © 2016 AAPG Pacific Section and Rocky Mountain Section Joint Meeting, Las Vegas, Nevada, October 2-5, 2016