The ARG (Acoustic Research Group) at the University of Canterbury has a goal to image objects buried under the seafloor. In order to image these objects, accurate models and a good understanding of the acoustic parameters of the seafloor are needed. The relevant acoustic parameters to model the seafloor include: physical classification, speed of sound, acoustic impedance, density, and roughness.
Acquiring accurate acoustic parameters for the seafloor is a challenging task that requires different approaches depending on the environment, especially in locations
with low visibility. This thesis looks at a non-invasive method of acoustically examining the roughness spectrum and acoustic impedance of the seafloor. To achieve these goals, spherical transducers with a continuous bandwidth between 30 kHz and 130 kHz were chosen to examine the frequency dependent specular and diffused components of the reflection from rough seafloor surfaces.
Using spherical transducers allows the surface scattering to be examined independently of the transducer beam-pattern. This thesis examines the range and validity of the tools required to make these measurements and presents empirical results measured in the controlled environment of a cylindrical 3.5m diameter by 2m deep sonar test tank. Using results from the sonar test tank, this thesis demonstrates that the acoustic impedance, speed of sound, and density of the material of the seafloor can be measured from a single seafloor realisation measured at multiple angles, provided the seafloor material has a critical angle. Additionally, this thesis demonstrates that the spectrum of the specular and diffused components of the rough-surface reflection requires averaging multiple seafloor realisations to be statistically relevant. As well as describing the benefits of using spherical transducers, this thesis presents some of the problems and potential solutions for future applications.
| Identifer | oai:union.ndltd.org:canterbury.ac.nz/oai:ir.canterbury.ac.nz:10092/4242 |
| Date | January 2010 |
| Creators | Noonchester, Mark |
| Publisher | University of Canterbury. Electrical and Computer Engineering |
| Source Sets | University of Canterbury |
| Language | English |
| Detected Language | English |
| Type | Electronic thesis or dissertation, Text |
| Rights | Copyright Mark Noonchester, http://library.canterbury.ac.nz/thesis/etheses_copyright.shtml |
| Relation | NZCU |
Page generated in 0.0036 seconds