Bayesian inversion of propagation and reverberation data

A Bayesian matched-field inversion approach to infer geoacoustic and scattering properties of the seabed is applied to simulated propagation and reverberation data received on a towed horizontal array. The approach is based on the method of fast Gibbs sampling (FGS) of the posterior probability dens...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of the Acoustical Society of America 2006-11, Vol.120 (5_Supplement), p.3356-3356
Main Authors: Nielsen, Peter L., Dosso, Stan E.
Format: Article
Language:English
Subjects:
Marine
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A Bayesian matched-field inversion approach to infer geoacoustic and scattering properties of the seabed is applied to simulated propagation and reverberation data received on a towed horizontal array. The approach is based on the method of fast Gibbs sampling (FGS) of the posterior probability density to estimate uncertainties in both geoacoustic and scattering parameters for broadband acoustic data in realistic shallow-water environments. The FGS is linked to an acoustic propagation model that simultaneously provides complex acoustic pressure at short propagation ranges and long-range reverberation intensity. The inversion algorithm is initially applied to long-range reverberation data alone to assess the geoacoustic information content of reverberation in terms of marginal posteriori probability densities for the environmental parameters. A reduction in uncertainty for the extracted geoacoustic and scattering parameters is demonstrated by a simultaneous inversion of the propagation and reverberation horizontal array data.
ISSN:0001-4966
1520-8524
DOI:10.1121/1.4781454