Integrated Rock-Typing Workflow in Shallow Unconventional Reservoir in Kuwait

Abstract Heterogeneous lithofacies distribution resulting into a complex rock-type model in shallow unconsolidated reservoir has a direct role on fluid distribution and trapping mechanisms. A systematic evaluation of these rock-types is necessary for proper reservoir characterization and modeling. In reality, the lithofacies leading to rock-types act as the building blocks to construct a realistic static model, which serves in the understanding of the dynamic behavior of the reservoir. During this study, 202 wells were selected across the field to capture the vertical and lateral heterogeneity of the reservoir, out of which 93 wells have cores. During a first step, a lithofacies prediction model was created from the core sedimentological description, X-Ray Diffraction (XRD), and wireline logs (raw and mineralogical logs) using probabilistic classification schemes. In a second step, petrophysical data like Routine Core Analysis (RCAL), Mercury Injection Capillary Pressure (MICP), were included to build rock-types associated with the different lithofacies. This integration workflow has resulted in a robust lithofacies and rock-type model consisting of nine lithofacies and five rock-types respectively. It was also noticed that silty non-pay and marginal pay reservoir have inadequate MICP data. Subsequently, two wells were selected and MICP data will be collected for improved and more confident modelling in future. This model assists to predict lithofacies and rock-types in un-cored wells provided a set of relevant logs are available. The integrated workflow ensures that the lithofacies and rock-types determined at the wells are consistent all over the study area. The identified lithofacies and rock-types will add great value in building a realistic reservoir static model since they are able to explain the fluid distribution pattern and the concept of barriers and baffles in the reservoir. This will also assist in optimized perforation and completion plans for the reservoir. Ultimately, the input data are readily available for future field-intensive reservoir characterization.