Focalização aplicada à tomografia acústica por campo casado em um ambiente de águas rasas

Abstract

Matched field tomography (MFT) is an inverse problem-solving method used to estimate a sound speed profile within a specific volume of ocean water by exploit- ing the complete acoustic field structure produced by a source in this environment. Its multidimensional solution heavily relies on the accuracy of input data available on bottom properties and problem geometry (source and receiver positioning, and bathymetry data) to avoid acoustic model mismatch errors. To overcome such chal- lenges and reduce the computational cost of MFT, the technique of focalization, an optimization and global search method, can be employed. This study applies and tests the robustness of focalization as a means to improve MFT results by absorbing input data imprecision and reducing computation time. For this pur- pose, a genetic algorithm is implemented and utilized on a dataset obtained from an experiment conducted in shallow waters, where the source position and bottom slope vary over time. The focalized MFT with geometry adjustment exhibited a significant improvement of 75% in inversion results and 50% reduction in processing time, on average, compared to classic MFT. The obtained outcomes highlight the applicability of focalized MFT as an alternative and complementary tool to sound speed measurements performed by oceanographic equipment, thereby contributing to ocean monitoring and enhancing our understanding of oceanic processes and their implications on the dynamic nature of the marine environment.