A Review of Current Techniques for Determination of Water Saturation From Logs

Abstract The basic saturation and log response equations are reviewed. It is concluded that conventional saturation calculations account for lithology and rock-type changes, but that they are susceptible to uncertainties in water resistivity (Rw), true resistivity, porosity and cementation factor (m). It is shown that resistivity-apparent porosity plots are useful in wells with minimum petrophysical data. Knowledge of Rw, m, the slope of the sonic log-porosity relation or the slope and intercept of the neutron log-porosity relation are not necessary, provided they are constant. Advantages and limitations are illustrated with examples. It is concluded that Rwa -depth plots are useful where Rw is unknown and lithology varies, provided m is known for all lithologies involved. Ros -So relations may be useful for determining water saturation when only a few porous intervals of constant rock type are present, and when either Rw or formation factor (but not both) are unknown. Finally, an example is reviewed to illustrate that, often, no one of the above techniques by itself may be diagnostic, and also to emphasize the need to utilize all available data. Introduction The determination of fluid saturations is still one of the prime functions of the petrophysical engineer. Although this problem has been continuously faced in day-to-day evaluation work since the advent of petrophysics, it still presents technically challenging problems. It is realized that hydrocarbon saturation is the quantity of real interest. However, with few exceptions, the problem resolves into determination of water saturation as defined by the following relationships: ....................(1) ...................(2) ....................(3) where = the fractional part of the pore volume filled with water of resistivity Rw = resistivity index = saturation exponent = true formation resistivity = formation resistivity factor* = fractional porosity = cementation exponent. Historically, the approach to this problem has been to determine resistivity index 1 from borehole measurements, and from 1 to calculate Sw using either an assumed value for n or one established from laboratory experiments. A discussion of the validity of laboratory determined values of n is beyond the scope of this paper. It will be assumed that the appropriate value for n is known, and this paper will discuss recent experience with the following techniques for determining 1:conventional saturation calculations,Ra vs plots,Rwa plots, andSo vs Ros relations. Conventional Saturation Calculations The time-honored process for making water saturation calculations involves the following steps:porosity is obtained from a core or a porosity log (sonic, neutron or density log);formation factor is calculated from Eq. 3 using an estimated m or one obtained from laboratory measurements or from resistivity measurements in 100 per cent water-bearing intervals;1 is calculated from Eq. 2 using a true resistivity Rt obtained from an appropriate resistivity device F, as calculated from Eq. 3 and an estimated Rw or one obtained from a water recovery in a nearby zone or another well, or one calculated from the SP log; andSw is calculated from Eq. 1 using the 1 calculated from Eq. 2 and n. This technique has the advantages of being well established and, therefore, relatively easily discussed with management and other log analysts. JPT P. 1425ˆ