Abstract

This study presents a hybrid optimization framework that combines Simulated Annealing (SA) with the Quasi-Newton Method (QNM) for seismic inversion to estimate acoustic impedance from post-stack seismic data. SA offers robust global search capabilities, while QNM ensures fast local convergence. By integrating these two complementary techniques, the proposed method addresses the limitations of each and enhances inversion accuracy and efficiency. The hybrid approach was applied to real seismic data from the Blackfoot field in Canada. Results demonstrate significant improvements in resolution and accuracy, with a strong correlation (0.99) between the inverted and real well log impedance, and a reduced RMS error of 0.26. Statistical measures including mean, mode, standard deviation, and correlation confirm the reliability of the method. Inverted sections reveal a distinct lowimpedance anomaly (8000–8500 m/s*g/cc) between 1045–1060 ms two-way travel time, interpreted as a potential sand channel. These findings highlight the effectiveness of the hybrid SA–QNM approach in high-resolution reservoir characterization.