Lithology Prediction Using Wave-Equation-Based AVO Seismic Inversion in Complex Geological Settings of the Pannonian Basin

The research area within the Pannonian Basin presents a complex geological setting, intricately divided into four distinct lithostratigraphic units. These units share similar lithological-petrophysical characteristics and respond to well-log measurements. In this case study, we delve into the application of Wave Equation-Based AVO (WEB-AVO) for precise lithology prediction, on an onshore oil reservoir within a sandstone-conglomerate formation. The WEB-AVO algorithm represents a seismic inversion scheme wherein the elastic wave equation is iteratively solved. Notably, it accounts for multiple scattering, mode conversions, and transmission effects across the target interval. The key steps of the WEB-AVO approach are initial elastic property estimation, wave-equation deployment and scattering effects, and direct solving for compressibility and shear compliance. A distinctive feature of WEB-AVO lies in its direct solution for compressibility (K, the reciprocal of bulk modulus) and shear compliance (M, the reciprocal of shear modulus). Unlike conventional AVO inversion techniques that focus on impedances, this approach provides more relevant information for quantitative interpretation. For the lithology prediction, we used a Bayesian classification algorithm for build facies volumes. Classification was done based on Kappa and M data in 6 wells, both absolute and contrast (relative) as input data. These results underwent significant analysis to validate their reliability. The result of the prediction illustrates very good match between the measured and predicted lithology logs for wells. Results of our research showcase the efficacy of employing elastic properties derived from WEB-AVO inversion for lithology prediction within a sandstone-conglomerate oil reservoir in the Pannonian Basin. Rather than dismissing observed non-linear events as mere noise, we recognise their significance. These unconventional seismic responses provide valuable additional information. By incorporating these events into our workflow, we enhance our understanding of subsurface properties. WEB-AVO demonstrates its predictive robustness beyond well locations. Elastic property predictions extend away from wells, enabling comprehensive reservoir characterisation. Notably, this method achieves efficiency gains by eliminating the need for specialised preconditioning—unlike traditional linear inversions. In summary, the integration of WEB-AVO inversion offers a powerful tool for lithology prediction, even amidst complex seismic data.