A Method of Predicting Saturation Exponents in Logging

Abstract

Abstract Water saturation is calculated from resistivity logs by using the relationship where n is the saturation exponent. From this equation, it is evident that in order to calculate accurate water saturations from logs, the value of n must be known. In this research project, saturation exponents and electrical formation factors were measured using three different salinity waters on 63 core samples from two wells in the same field. Porosity and permeability were also determined on each of the cores. The log readings from the SP Log, Density Log, Gamma Ray. Log, and Sidewall Neutron Log were read to coincide with each core sample. The lithology of samples ranged from clean to shaly fine grained sandstone. Both the saturation exponent and formation factor were found to be greatly affected by the saturating water resistivity. The three waters used had resistivities of 0.06, 0.2, and 2.0 Ohm-Meters. As the water resistivity increased, the saturation exponent and formation factor decreased. The saturation exponent averaged about 2.03 for the 0.06 Ohm-Meter water but decreased to an average of about 1.21 for the resistive 2.0 Ohm-Meter water. All of the core analysis and well log data was used in a regression analysis computer program to predict saturation exponent. Saturation exponent was predicted with the computer program to an average error of about 9 percent. Water resistivity was by far the most percent. Water resistivity was by far the most important independent variable in the prediction equation and permeability was second. By using water resistivity by itself, the saturation exponent was predicted with an average error of 15 percent. This research shows that in either shaly sand or formation with high water resistivity large errors in determining water saturation can be made by assuming that the saturation exponent is equal to 2. This work further shows that saturation exponent can be predicted from routine core analysis and well log data.