Reservoir Characterization Of Fractured Reservoirs In Western Canada

Abstract

Abstract The author has worked on a number of fractured reservoirs in Western Canada, which show common characteristics. Production performance, pressure transient responses and stimulation results are discussed. A reservoir characterization is presented which is consistent with observed production performance, pressure transient responses, production logging results, core analysis and well stimulation. A key component is structural geological style. This description has been applied to a number of different reservoir situations. It has application to stimulation design, predicting reservoir performance, numerical simulation and pressure transient analysis. An example is also highlighted from a gas condensate reservoir. Introduction The material in this paper was derived from a number of large detailed reports (for example see reference 1). Such detail and volume of material cannot be covered in a single technical paper. The paper therefore only presents a summary of a number of key concepts derived based on the author's experience. Evaluating a fractured reservoir [after Nelson(2)] involves four main steps:Interpreting the origin of the fracture system. This information allows one to predict geometry and the extent of communication.Determining petrophysical properties of the fractures and matrix. This allows for prediction of the variation in reservoir response. The relative storage (ie porosity) must be determined as well as effective permeabilities. Another important property is compressibility.The flow interaction between the matrix and fracture system is evaluated to determine ultimate reserves from the reservoir.Classification of the reservoir. Depending on the type of flow interaction the reservoir will fit one of several depletion strategies. Note that most of the variations in strategy apply to waterflooding oil reservoirs. A large proportion of the above data is obtained from core. The following conventional core analysis data and plots can be used:Core permeability vs. core porosity-all data.Core permeability vs. core porosity-sorted by lithology.Vertical permeability vs. horizontal permeability.K90 plotted vs. Kmax. Fractured reservoirs do not show the typical straight line relation on core permeability vs. core porosity (semilog) plots. Typically lower porosity rock is more prone to fracturing. Fractured reservoirs also tend to have higher anisotropy, which is seen as large variation in K90 vs. Kmax. Core Description In the author's experience, based on a number of studies in Western Canada, there is a typical core permeability vs. core porosity relationship. The following diagrams and points are based on an example from the central Foothills:Core permeability vs. core porosity is shown in Figure 1. The data show tremendous scatter, which is typical of a fractured reservoir. There are a total of 762 data points, which have an average air permeability of 23.392 mD and an average core porosity of 4.11%. At first glance the data looks almost completely random. However, more careful examination shows there is a triangular distribution. As outlined earlier, lower porosity core is more prone to fracturing. The lower bound of the triangle approximates matrix properties.