Formation waters in hydrocarbon reservoirs can have complex chemical composition. In the high-salinity hydrocarbon reservoirs of the Middle East, formation waters contain, in addition to sodium and chlorine, substantial concentrations of calcium, magnesium, potassium, strontium, carbonate, sulphate and bicarbonate ions. In recent times, advances in drilling fluids chemistry have resulted in mud-filtrate composition very different from the traditional NaCl brines. Mud filtrate can include high concentration of KCl, NaBr, CaCl2 and CaBr2. The presence of the different elements impacts the response of the logging sensors to the presence of formation water and mud filtrate in the formation. It has been customary to account for the conductivity of the different ions by converting their concentrations to equivalent NaCl salinity. While charts for such conversions have existed, their systematic use has been hampered by the lack of user-friendly software. Furthermore, while this approach might satisfactorily account for the conductivity of the brine, very large errors occur when it is used to estimate the water-response parameters for different nuclear logging sensors.
Through the use of several software algorithms, we present a workflow for the correct modeling of the log responses of complex brines. The workflow is applied to the data of a well drilled using water-base mud through a Jurassic carbonate gas-bearing formation. The use of an equivalent NaCl salinity resulted in choice of salinity values not supported by the measurements. Furthermore, not all log responses could be correctly modeled. The application of our workflow using measured mud and formation water composition yielded excellent reconstruction of measured logs, and the petrophysical outputs matched core results very well. The workflow allowed us to quantify the errors introduced when equivalent salinity was used to estimate the downhole properties and log-response parameters of the brine.