Formation Evaluation in Thin Sand/Shale Laminations

Abstract Formation evaluation in thin sand-shale lamination seeks first to determine sand resistivity, volume fraction, and porosity. Afterwards, saturation and volume are simple Archie applications. Resistivity anisotropy techniques can provide estimates of sand resistivity and volume fraction, but good results depend on the choice of the anisotropic shale point. The same shale point should be used in the determination of sand porosity. Difficulties will arise when anisotropy is not caused by sand-shale laminations, when no sand-shale point exists, or when the nearby thick sand-shale is not representative of the sand-shale in the laminations. In producing fields that have undergone several waterfloods, water resistivity is often unknown in the swept thick sands and might not be representative of the water in the unswept thin sands. As discussed previously, NMR offers useful insights into the petrophysics of thin sand-shale laminations. Typically, 1D high-resolution data is acquired to estimate sand volume fraction, porosity, and permeability, and 3D fluids data is used to evaluate the hydrocarbon type and content in the thin sands. However, shallow depth of investigation, slow logging speed, and sometimes unfavorable signal-to-noise ratios limit the applicability of the NMR technique. In this paper we first demonstrate the variability of sand resistivity, volume fraction, and porosity output depending on the input parameters. Next, we show the complementary aspects of the resistivity anisotropy and NMR techniques. Since several are the same or determined independently, using both datasets ensures more plausible results in thin beds than either stand-alone technique could provide. Field examples of a straightforward case with a well-defined anisotropic shale point, as well as a difficult case with multiple shale points, are used to demonstrate the new workflow. In general we see improvements in the estimation of the hydrocarbon in place; however, not all thin beds offer large hydrocarbon volumes. Introduction The topic of formation evaluation in thin sand/shale laminations has been treated by many authors for the last 30 years. In recent years, we have studied, experimented, applied and refined the interpretation technique for thin sand/shale laminations using both NMR and triaxial induction (3D induction) data on numerous occasions. To keep the paper readable, we have broken the topic into three parts. First, we discussed the NMR petrophysics in thin sand/shale laminations (Cao Minh and Sundararaman, 2006).1 Next, we presented a graphical method to analyze resistivity anisotropy in thin sand/shale formations (Cao Minh et al., 2007).2 Both papers contained many useful references that will not be re-quoted here. This paper is the third and last in the series. We will show the thin beds workflow using both 3D NMR and 3D induction data. Combining the two dataset provides useful check points to ensure the best possible interpretation in thin sand/shale formations. A quick review of 3D induction technique is discussed first. The important point is to understand how sand resistivity and sand volume fraction are derived and the effects of the input parameters on the results. Next, we show how to use NMR to verify these two outputs. The third verification is the sand porosity computation. Finally, the fourth and last verification is the volume of hydrocarbon. Although crossvalidation between 3D induction and NMR interpretation techniques does not guarantee an accurate evaluation of reserves, it does indicate that the results are plausible. What we seek first is to be able to say yes, hydrocarbon is indicated by both tools, or no, hydrocarbon is indicated by neither tool. The case to be avoided is having hydrocarbon indication by only one tool. In the latter, we have found that the cause is likely in the choice of input parameters. In many cases, a "yes/no" answer is as important as a "how-much" answer because it allows the operator to test or to abandon the well. Finally, we use imaging logs/core data to confirm the presence of thin beds.