Multi-step inversion of seabed geoacoustic and scattering parameters using joint propagation and reverberation data

Inferring seabed geoacoustic and scattering parameters from acoustic data is challenging due to inherent parameter correlations. Combining short-range propagation with long-range reverberation data helps mitigate these challenges, especially when multi-frequency datasets are used to address local optimization issues and parameter sensitivities. However, using the full dataset can lead to high computational costs and reduced parameter sensitivity across frequencies. This paper proposes a multi-step joint inversion method that sequentially updates parameters based on their frequency-dependent sensitivity, ensuring convergence to the global optimum while reducing computational demands. This approach enhances the efficiency and reliability of seabed parameter estimation, offering a promising solution for marine geophysical applications. The multi-step strategy is validated using both simulated and experimental data.