Abstract
Permeability is one of the most important petrophysical parameters in formation evaluation and reservoir description. Unlike porosity and saturation, permeability can be a non-zero rank tensor. Most published log-based permeability models are only used to determine scalar or isotropic permeability. However, more and more oil/gas reservoirs have been found in anisotropic formations. In these geological environments, the permeability, resistivity, and some other petrophysical parameters are frequently anisotropic. From an industry-wide viewpoint, very few logging tools provide measurements that can be used as inputs for the calculation of log-based permeability anisotropy (horizontal and vertical permeability). Multicomponent induction (MCI) tools are examples that can measure resistivity anisotropy. This paper describes the algorithms and interpretation workflow that can be used to assess permeability anisotropy from an integrated interpretation of resistivity anisotropy and conventional log-derived permeability.
As an extension of present permeability models in isotropic formations, a new permeability model in an anisotropic formation is obtained. Based on this anisotropic model, a new relationship between resistivity anisotropy and permeability anisotropy is determined in anisotropic formations. This relationship shows that the permeability anisotropy is a function of the resistivity anisotropy ratio and pore structure parameters. By comparison, the previously published results include only a few special cases in the new relationship. Assuming that the log-derived permeability data are available and are calibrated based on the effective permeability or one component of the permeability tensor, the remaining permeability components can be obtained from the relationship of resistivity and permeability anisotropy. In thinly laminated shale-sand formations, log-derived permeability and horizontal or vertical permeability cannot accurately represent true reservoir permeability because of the vertical-resolution limitation of logging tools. To overcome this limitation, the true reservoir permeability is evaluated from the calculated horizontal and vertical permeability based on a multimodal permeability tensor model. For practical applications in anisotropic reservoirs, an interpretation workflow is presented for permeability anisotropy evaluation with the joint interpretation of resistivity anisotropy and log-derived permeability from conventional/advanced sensor logs (e.g., resistivity, imaging, and sonic).
The algorithms and workflow are validated by using the numerical simulation and field data. The new workflow has been used in the permeability anisotropy interpretation of synthetic data with and without random errors. After the synthetic data validation, the workflow is applied to field log interpretation. Both applications showed that this new addition of the permeability anisotropy should significantly assist in the accurate assessment of the reservoir, as well as in fracture detection and subsequent oil development and production.
