Abstract

This study presents a comparative evaluation of genetic algorithm (GA) and artificial gorilla troops optimization (AGTO) for post-stack acoustic impedance inversion using seismic data from the Blackfoot field, Canada. AGTO, a novel global optimization method, was developed and applied to improve inversion accuracy, resolution, and efficiency. Comprehensive analysis across time, frequency, and spatial domains demonstrates that AGTO consistently outperforms GA. AGTO achieves higher correlation with well-log data (Avg CC = 0.96), and clearer resolution of geological features, including anomaly zones. Crossplot analysis shows stronger agreement with well-log impedance (R² = 0.84 for AGTO vs. 0.71 for GA), and statistical metrics such as mean, median, standard deviation, and range more closely match the well data in AGTO results.. Additionally, AGTO converges faster, reduces inversion error more effectively, and completes the full seismic inversion in just 32 hours, compared to 67 hours for GA. Furthermore, statistical comparison confirms AGTO delivers more geologically consistent and accurate impedance estimates than GA, with better alignment to well-log-derived impedance values. Overall, AGTO proves to be a more accurate, robust, and computationally efficient inversion technique, offering significant advantages for high-resolution subsurface characterization and reservoir characterization