Shelf Break Acoustics: Theory, Computation, and Statistics

Abstract

The goal of this proposal is to contribute to a broad Task Force Ocean investigation thataddresses, in shelf break regions, how oc"ean processes can quantitatively affect acousticpropagation, how inversion techniques can best estimate environmental and acousticparameters, and how acoustic source and ocean feature localizations can be improved. Thisproposal has three objectives and task areas, each motivated by requirements to analyzeresults for the interdisciplinary New England Shelf Break Acoustic program; the latter wasconceived for TFO and will be carried out by PIs at the Woods Hole OceanographyInstitution. The first objective is to determine the sensitivity of acoustic quantities in termsof parameter variations of ocean feature models for the NESBA region. Such results willhave complementary advantages to those to be obtained using an advanced acousticsensitivity kernel from WHOI. The second is to construct a model suite for estimatinggeoacoustic parameters of mixed-constituent seabed sediments relevant to NESBA. Adistinctive feature is that the basic-physics seabed models specify parameters from relativelyfew physical measurements, and consequently will provide new and efficient inversionresults from NESBA data. The third is to improve accuracy and efficiency of ocean andgeoacoustic parameter inversions by constructing new expressions for noise components ofthe data covariance matrix. A key requirement for these expressions is the feasibility fortaking advantage of machine learning techniques in evaluations involving large data sets."The outcomes from the three task areas will increase the usefulness of the NESBA resultsand conclusions for Navy applications.

Document Details

Document Type
DoD Grant Award
Publication Date
Sep 30, 2019
Source ID
N000141912636

Entities

People

  • William L. Siegmann

Organizations

  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Coastal Oceanography
  • Neural Network Machine Learning.
  • Neurological Diseases/Conditions/Disorders

Technology Areas

  • AI & ML