Inferring low-frequency compressional sound speed and attenuation in muddy sediments from long-range broadband sound propagation

Abstract

While researchers have successfully applied many geo-acoustic inversion methods, involving normal mode analysis or matched field processing, to invert seabed parameters in sandy bottoms, inferring low-frequency compressional sound speed and attenuation in mud is still a challenging problem, especially for a marine sediment where a mud layer overlies a sandy bottom. Recent studies show that there is an ambiguity between the low-frequency attenuation of mud and sand in three different inversion algorithms based on (1) modal amplitude, (2) transmission loss, and (3) spatial coherence measurements [JASA 143, 1798 (2018)]. An increase (decrease) of mud attenuation can be compensated by decreasing (increasing) the attenuation in the sandy basement. In this paper, an inversion method combining these three inversion algorithms and statistical inference techniques (e.g., Bayesian-Maximum Entropy, [JASA 138, 3563–3575 (2015)]) is utilized to analyze the long-range broadband acoustic signals measured by several vertical and horizontal line arrays during the Seabed Characterization Experiment 2017 conducted in the New England Mud Patch. Reliable estimates of the low-frequency mud properties are obtained by removing the ambiguity. Finally, the inverted results are compared with historical experimental data and theoretical predictions. [Work supported by ONR Ocean Acoustics.]While researchers have successfully applied many geo-acoustic inversion methods, involving normal mode analysis or matched field processing, to invert seabed parameters in sandy bottoms, inferring low-frequency compressional sound speed and attenuation in mud is still a challenging problem, especially for a marine sediment where a mud layer overlies a sandy bottom. Recent studies show that there is an ambiguity between the low-frequency attenuation of mud and sand in three different inversion algorithms based on (1) modal amplitude, (2) transmission loss, and (3) spatial coherence measurements [JASA 143, 1798 (2018)]. An increase (decrease) of mud attenuation can be compensated by decreasing (increasing) the attenuation in the sandy basement. In this paper, an inversion method combining these three inversion algorithms and statistical inference techniques (e.g., Bayesian-Maximum Entropy, [JASA 138, 3563–3575 (2015)]) is utilized to analyze the long-range broadband acoustic signals measured by several vert...