Enhancing hydrocarbon prospect delineation through artificial intelligence driven integration of seismic attributes and inversion in ‘OS’ field, offshore, Niger Delta

Several studies have employed multi-seismic attributes and seismic inversion to delineate potential hydrocarbon zones on a seismic scale. However, these methods often rely on individual attribute maps and their integration struggle to effectively harness the collective information embedded in their results, leading to hydrocarbon zones being bypassed. To improve precision and reduce uncertainty in integrating these methods, this study introduces an artificial intelligence-driven approach to incorporate results from seismic inversion and multi-attribute analysis for enhanced characterization of hydrocarbon prospects. Key seismic attributes, including instantaneous amplitude, amplitude envelope, and instantaneous frequency, known for their potential as direct hydrocarbon indicators, were employed. Also, post-stack seismic inversion was utilized to derive acoustic impedance, providing a quantitative measure of subsurface properties. To enhance the delineation of hydrocarbon prospects, a computer vision algorithm was designed and implemented using the OpenCV library in Python on the attribute maps to isolate zones corresponding to potential hydrocarbon zones based on their distinctive properties. This process isolated hydrocarbon-prospective zones within each attribute map. These enhanced zones were subsequently integrated using a computer vision algorithm designed to identify areas of overlap, indicating potential hydrocarbon prospects. The resulting integrated map yielded precise and accurate hydrocarbon prospect identification by ensuring alignment with the criteria defined by all employed attributes. The results precisely identify hydrocarbon-bearing zones by reducing uncertainty, demonstrating the effectiveness of integrating seismic attributes and inversion data using artificial intelligence. This innovative approach enhances hydrocarbon prospect evaluation by improving efficiency, accuracy, and precision in extracting and integrating critical information from seismic data. An offshore field (‘OS’) in the Niger Delta Basin was used as the study area.