The paper was presented at the SPE/DOE Unconventional Gas Recovery Symposium of the Society of Petroleum Engineers held in Pittsburgh, PA, May 16–18, 1982. The material is subject to correction PA, May 16–18, 1982. The material is subject to correction by the author. Permission to copy is restricted to an abstract of not more than 300 words. Write: 6200 N. Central Expwy., Dallas, TX 75206. Introduction Reliable evaluation of hydrocarbon resources encountered in shaly clastic reservoirs of low porosity and low permeability is an important although difficult task. Log-derived estimates of the volume, type and distribution modes of various clay minerals, determination of cation exchange capacity (CEC) and Qv (CEC per unit of total pore volume), and properly selected water saturation calculation models assist formation evaluation. Since shaly clastic reservoir rocks require extensive core sampling for CEC and Qv analysis, which is tedious, time consuming, and expensive, attempts have been made to correlate such CEC and Qv data with one specific or a combination of several well logging measurements. The latter include the spontaneous potential, gamma ray, natural gamma ray spectral data, dielectric constant and acoustic-, density- and/or neutron-derived porosity, etc. Constraints associated with these concepts will be reviewed. Discussed in this paper, is an innovative digital shaly sand evaluation approach (CLASS), which provides information on total and effective reservoir porosity, total and effective fluid distribution based on the Waxman-Smits equation, reservoir productivity, silt volume, and volumes, types and distribution modes of clay minerals present in subsurface formations. Both basic concepts and field case examples will illustrate this method. CLAY MINERALS Clay minerals, used as a rock and particle term, describe an earthy, fine-grained, natural material which develops plasticity when mixed with a small amount of water. Such clay minerals significantly affect important reservoir properties such as porosity, water saturation, and permeability. Clay minerals are composed of small crystalline particles which are classified according to their crystal particles which are classified according to their crystal structure. Important ones of interest to the petroleum engineer and geologist are kaolinite, montmorillonite, illite, chlorite, and mixed-layer minerals. They are essentially layered hydrous aluminum silicates which may contain small amounts of alkalies and alkaline earths and have some substitution of aluminum by other cations, such as magnesium, iron, etc. The most common clay minerals, their composition, matrix density, hydrogen index, CEC, and distribution of potassium, thorium, and uranium based on natural gamma ray spectral information are listed in Table 1.Numerous experimental data show that the CEC value of clays is directly related to their capacity to absorb and hold water. Clays of the montmorillonite (smectite) group have the greatest capacity to absorb water and also the highest CEC values. Kaolinite and chlorite have very low CEC, and their capacity to hold water is also low. Shales can be defined as an earthy, fine-grained, sedimentary rock with a specific laminated character. Based on the analyses of 10,000 shales Yaalon describes the mineral composition of the average shale as follows: clay minerals (predominantly illite), 59%; quartz and chert, 20%; feldspar, 81%; carbonates, 71%; iron oxides, 30%; organic materials, 1%; others, 2%. Generally speaking, illite appears to be the dominant clay mineral in most of the shales investigated. Chlorite mica is frequently present, smectite is a common component in Mesozoic and younger shales, and kaolinite usually occurs in small amounts only. Therefore, a typical shaly clastic reservoir rock and/or a typical shale formation may consist of several components. Hence, no universal shale parameter can be used to characterize a specific type of argillaceous sediment or rock. P. 67

Reservoirs with thin laminations can be more accurately evaluated by using logging tools with inherently better vertical resolution, by employing enhanced vertical resolution input processing methods, and by incorporating interpretation models that properly handle log inputs with different vertical resolutions and reconstruct all outputs with high vertical resolution. The paper discusses a specific high-resolution interpretation model and provides comparative analyses of how model outputs are affected by the vertical resolution of its input logs and by the reservoir type. Three field examples are provided. Increases in predicted hydrocarbons were noted in two of these examples when the input log resolution was increased. In these two examples, the observed increases were confined to a number of isolated thin beds. In the third field example, significant decreases in predicted hydrocarbons were observed when high-resolution input data were used; the reservoir in this case appears to illustrate thin sand-shale laminae extended over a 27 metre interval.

This paper waa selected for presentationby an SPE ProgramCommittee followingreview of Informationcontained in an abstract submittedby the author(a). Contentsofthe paper, aa presented,have notbeen reviewed bythe Societyof PetroleumEngmearsand are subjectto correctionbythe author(s).The material, es presented,does notnecessarily reflect any positionofthe Society of PetroleumEngineera,itaoffk?era, or members.Papers presentedal SPE meetingsare subjectto publicationreview by Editorialcommittees01the Societyof PetroleumEngineers Permissionto copy is restrictedtoan abstractof notmorethan 300 words Illustrationsmay not be copied. The abstract should contain conspicuous acknowledgment of where and by whom the paper IS presented. Write ABSTRACTalso assist in determining when satisfactory solutions can Advanced analytical methods and formulas have been be obtained from log data alone and when additional developed for two-and three-mineral solutions in complex information (core data) is needed.

Summary We developed a new method to monitor gasflooding of oil reservoirs. The method is based on computing two porosities: the base true porosity determined before gasflooding and monitor-apparent neutron porosity determined after gasflooding. The base and monitor porosities provide determination of a hydrogen index of the reservoir fluid after the flooding. The hydrogen index is then used to determine saturation of the flood agent after flooding and, consequently, saturation of water and oil after flooding. An example of quantitative analysis of base and monitor logs illustrates monitoring of CO2 flooding by use of the described method. Introduction The method of monitoring gasflooding presented in this paper is intended mainly for quantitative log interpretation at the well sites or remote computing centers. It requires a minimum of information about reservoir rock and fluid properties, most of which can be derived from the logs. The method is based on the computation of two porosities: the base porosity determined before gasflooding and monitor-apparent neutron porosity determined after the gasflooding. The primary disadvantage of this approach is that the monitor porosity has to be corrected for the excavation effect. The formulas proposed in Ref. 1 can be used for this purpose. However, these formulas take only the hydrogen index of fluid, porosity, and lithological composition of a reservoir into account. Specific flood agent, hydrocarbon types and gravity, different clay types, and accessory minerals are not taken into consideration. Thus, the correction for the excavation effect cannot always be accurate. If one desires higher-accuracy computations and if the specific chemical and physical properties of matrix, shale, formation water, hydrocarbons. and flood agent are known, one can use more comprehensive techniques for gasflood monitoring, such as the TMDFLOOD model described in Ref. 2. TMDFLOOD is a sophisticated log-analysis and modeling program more suitable for in-house, mainframe processing. It uses known chemical and physical properties of the reservoir, hydrocarbons. and the flood agent to model both the capture cross section, s, and ratio/porosity responses of a pulsed-neutron logging tool. Because this model is based on total neutron migration length and on direct ratio computations for the calculation of volume fractions of the flood agent, hydrocarbons, and formation water, it does not require a correction for the excavation effect. Higher accuracy of the monitoring method presented in this paper can also be achieved if chemical and physical properties of the reservoir and the flood agent are known. The neutron migration length and direct ratio calculations can be used to compute corrections to the monitor porosity and to derive corresponding formulas for the specific conditions. Once derived for a specific flood agent and a specific reservoir, these formulas can be used as part of the method presented here to correct monitor porosity for the "excavation effect" during the total life of the gasflood monitoring project. Refs. 3 through 6 also consider the quantitative monitoring of gasflooding by use of logging methods.