Inversion for geoacoustic model parameters in complex shallow water environments

Abstract

Hz). This research was partially successful in developing an inversion for generating estimatesof sound attenuation in sediments at the Shallow Water ~06 (SW06) experimental site. Theresearch proposed here extends the research to investigate performance improvements with newformulations of the warping operation and the sequential inversion method. The underlying goalof the work is development of a method using modal amplitude information from single sensorbroadband data at relatively close range for inverting attenuation of sound in various types ofmarine sediment.Most if not all of the previous work on geoacoustic inversion has been based on the visco-elastictheory of sound propagation in sediments. Inversions that adopted this approach have generallyfollowed Bayesian me"thods and used sound speed, density and attenuation as geoacoustic modelparameters in global search algorithms. Although the sound"" speed is a sensitive parameter, it isnot clear that this parameter set is the best one to use. This work proposes to use a morefu""ndamental parameter set consisting of physical parameters, such as porosity and grain contactstress, which define the material prop"erties of the sediment. These parameters are basic totheories such as Buckingham~s viscous grain shearing theory of sound propagation in marinesediments. The theory provides analytic relationships between the physical parameters and theacoustic quantities of so"und speed, attenuation and density. The use of Buckingham~s soundpropagation theory will be tested in simulations of inversions of" the modal time-frequencyinformation.All of the proposed aspects of the research plan will make use of experimental data from theSW06 experiment to study the type of broadband sound source that is best suited for use in the2017 ONR Seabed Characterization experiment. Use of experimental data from SW06 allowscomparison of results from the inversion of warped data with results from other techniques thatwere used.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 07, 2017
Source ID
N000141712659

Entities

People

  • Norman Chapman

Organizations

  • Office of Naval Research
  • United States Navy
  • University of Victoria

Tags

Readers

  • Acoustical Oceanography.
  • Data Mining and Knowledge Discovery.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference