Aspects of Multicomponent Resistivity Data and Macroscopic Resistivity Anisotropy
- J.H. Schoen (Joanneum Research) | Liming Yu (Baker Atlas) | D.T. Georgi (Baker Atlas) | O. Fanini (Baker Atlas)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Evaluation & Engineering
- Publication Date
- October 2001
- Document Type
- Journal Paper
- 415 - 429
- 2001. Society of Petroleum Engineers
- 4.3.4 Scale, 1.6 Drilling Operations, 5.6.1 Open hole/cased hole log analysis, 3.3.2 Borehole Imaging and Wellbore Seismic, 1.12.2 Logging While Drilling, 5.7 Reserves Evaluation, 4.1.5 Processing Equipment, 2.4.3 Sand/Solids Control, 5.5.2 Core Analysis, 1.14 Casing and Cementing, 5.1.5 Geologic Modeling, 1.2.3 Rock properties, 5.2 Reservoir Fluid Dynamics, 5.1.1 Exploration, Development, Structural Geology, 1.10.1 Drill string components and drilling tools (tubulars, jars, subs, stabilisers, reamers, etc)
- 1 in the last 30 days
- 447 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 12.00|
|SPE Non-Member Price:||USD 35.00|
New induction-logging hardware makes it possible to obtain both resistivity and resistivity-anisotropy data. Resistivity anisotropy, the ratio of vertical to horizontal resistivity, is the macroscopic effect of thinly layered formations in which logging tools have insufficient vertical resolution to properly resolve the individual beds, or laminae. Generally, there are two special types of layering.
Laminated shaly sands. The sediment consists of thin-bedded sand/shale sequences (anisotropy in these sands originates from the contrast of shale and sand resistivity).
Finely layered anisotropic sands. The sand is composed of layers of different grain sizes/sorting (anisotropy in these sands originates from the resistivity contrast associated with the different water saturation).
The interpretation of conventional resistivity data is well understood. However, interpretation of vertical resistivity and resistivity anisotropy is not well understood and is often counterintuitive. We have used forward modeling to illustrate the effects of porosity variability in layered formations. For example, we have investigated the porosity-layering effect, which varies from an average porosity of 20 p.u. by +/-5, 10, and 15 p.u. Three types of layering were considered: graded bedding and square- and sin2-porosity variation with depth. The modeling shows that sharp bed boundaries create the maximum resistivity anisotropy for any two component resistivity distributions.
The new induction-logging hardware comprises three mutually orthogonal, transmitter-receiver coil configurations that measure all data necessary to derive both resistivity and resistivity anisotropy of the formation in vertical, deviated, and horizontal wells. Simple models illustrate the physics, and the tools' capabilities are demonstrated with a synthetic example. Based on the petrophysical analysis of porosity contrast and layering type, a resistivity model is constructed, and the tool responses of this model are computed. With inversion techniques, both resistivity and resistivity anisotropy can be recovered.
Evaluation of thinly laminated reservoirs is critical in today's deepwater plays. In most deepwater plays (e.g., the Gulf of Mexico, west of Africa, and offshore Brazil), thinly laminated turbidite sands are associated with large hydrocarbon accumulations. Because deepwater development costs are high, it is essential that all reserves be identified and accurately determined. However, these laminated-sand packages are difficult to identify and to evaluate accurately with conventional, coaxial induction-tool data. Newly developed, multicomponent induction hardware facilitates the accurate evaluation of such thinly laminated sand/shale sequences. Further, recognition of the presence of thin, clean sand lamina is important because laminated sands are much more productive than uniform, dispersed shaly sands.
Generally, when a potential reservoir containing thinly laminated sands and shales is recognized, petrophysicists and log analysts correct the low-resistivity induction-log data for thin bed effects. Once they correct the data, they calculate saturations and compute reserves based on one of the many shaly sand evaluation equations.1,2 Even then, however, it is not uncommon to underestimate the hydrocarbon volume and bypass pay.
Klein3-5 recognized that conventional induction-log resistivity underestimated the resistivity of thin-sand components in a laminated sand/shale sequence. Even though the thin, individual sand and shale laminae may be isotropic, the layered sequence of sands and shales appear anisotropic on the scale of the logging tool measurements. Klein,4 Klein et al.,6,7 and Hagiwara8-10 derived averaging expressions for the macroscopic vertical and horizontal resistivities.
Herrick and Kennedy11 noted the anisotropic effects on resistivity in deviated wells and demonstrated the nonlinearity of the saturation exponent as the result of a trimodal, layered, anisotropic-sand model. Petrophysical evaluations of thin-bedded, laminated reservoirs are incorrect if traditional scalar saturation equations are employed. The phenomenon of macroscopic anisotropy leads to a series of complications and errors if traditional interpretation techniques are employed.7
Mollison et al.12 and Schoen et al.13 have introduced a new tensorial petrophysical model and procedure to evaluate anisotropic laminated sands. The following two-step analysis approach was proposed.
The decomposition of the sequence into two components-bulk shale and sand.
The petrophysical interpretation of the individual components based on the determined resistivity and volume fractions to calculate reservoir properties (e.g., porosity and saturation) for each component.
To use the method introduced by Mollison12 and Schoen,13 it is necessary to have both horizontal and vertical resistivity data (RH and RV, respectively).
Before the introduction of the multicomponent induction-logging instrument (3D ExplorerSM, 3DEXSM)*, only limited resistivity-anisotropy data were available from logging-while-drilling (LWD) data interpretation.14 The new multicomponent, induction-logging instrument, 3DEX, provides three mutually orthogonal measurements sensitive to resistivity anisotropy. To date, it has been successfully logged in vertical, deviated, and horizontal wells. Generally, in vertical wells, data from the coaxial coils (Hzz) are similar to conventional induction data and are straightforward to interpret. However, interpreting resistivity data from the coplanar coils (Hxx and Hyy) is much more complicated.
In addition to hardware issues, water sands are generally assumed to be homogeneous and isotropic. Thus, anisotropic water sands were unexpected until an initial 3DEX log showed a significant anisotropy (the upper picture of Fig. 1). Laminated shaly sands have, as expected, distinctly higher anisotropy (the lower picture of Fig. 1).
From a petrophysical perspective, it is straightforward to compute the vertical and horizontal resistivity components; however, the new measurements are made from within the borehole with a tool of finite dimension at set frequencies. Thus, it is necessary to consider the coil spacings, the conductivity of the formation (skin effect), and the borehole effects. These effects are accounted for with the tool's forward-modeling inversion procedure, which is discussed later.
|File Size||2 MB||Number of Pages||15|