The evaluation of carbonate reservoirs is a complex task because of the inherent heterogeneities that occur at all length scales. Rock properties may be defined differently at different scales and this introduces a challenge in capturing heterogeneity in a single rock volume. Heterogeneities at smaller length scales must be upscaled into larger scale volumes to better predict reservoir performance. The objective in this study is to define carbonate rock types at multiple scales and then upscale those rock types and associated properties to the whole core level.
Representative core plugs were selected in a heterogeneous reservoir interval based on statistical distribution of litho-types in the core. The litho-types were described by porosity and mineralogy variations along the core length using advanced dual energy XCT imaging. Plug-scale rock types were defined on the basis of petrophysical data and geological facies. High-resolution micro to nano XCT images were integrated in the rock typing scheme. Those rock types were upscaled to the whole core level by linking the core litho-types to the plug data.
The core litho-types (porosity and mineralogy) gave good representation of the whole core heterogeneity and were reliable for selecting representative samples. This allowed establishing the link between plugs and whole cores and hence upscaling rock type information to the whole core scale. The high-resolution digital images emphasized the different pore geometries in the samples and improved the definition of the rock types. Accurate porosity and permeability logs were derived along the core length and gave very good match with the plug data. Multi-scale porosity-permeability trends were investigated and found to have direct impact on the determination of upscaled permeability log at the whole core level.
The paper presents an advanced and quick tool for representative sample selection and statistical core characterization in heterogeneous reservoirs. The identified rock types at multiple scales provided new insights into carbonate heterogeneity and gave upscaling options for rock types and petrophysical data. The upscaled rock types at the whole core level enhance the prediction of dynamic imbibition data along the reservoir column for improved reservoir performance.