AI‐Enhanced MicroCT Analysis of Pore Systems in a Coquina With Boundstone Structure From the Morro do Chaves Formation, Sergipe‐Alagoas Basin, Brazil

The study of coquina samples is critical for the oil and gas industry, as these carbonate rocks often serve as significant reservoirs due to their unique porosity and permeability characteristics. Understanding the microstructural properties of coquinas, such as pore connectivity and crystal distribution, is essential for optimizing hydrocarbon recovery. This study employs MicroCT imaging and AI‐based segmentation to analyze a coquina plug (2‐SMC‐2AL) from the Sergipe Basin, collected at 96 m depth. Porosity was estimated through binarized 3D image analysis, using voxel and morphometric coefficients. A novel AI‐driven segmentation technique enhanced pore space identification, improving accuracy in quantifying porosity (13.81%) and pore density (0.1817/mm 3 ). Additionally, fluid flow simulations were conducted, revealing a permeability of 245 mD, consistent with the observed high pore connectivity. The sample exhibited selective intercrystalline porosity, where porosity increases with crystal thickness. Compositionally, the coquina consisted of 20% linear layers and 80% plane‐parallel boundstone. The integration of AI segmentation and flow simulations provided deeper information into the petrophysical properties of coquina reservoirs, demonstrating the potential of advanced imaging and computational techniques in reservoir characterization.