Date of Award

Spring 2023

Project Type

Thesis

Program or Major

Ocean Engineering

Degree Name

Master of Science

First Advisor

Anthony P Lyons

Second Advisor

Thomas Lippmann

Third Advisor

John E Hughes Clarke

Abstract

Synthetic aperture sonar (SAS) is based on synthetic aperture radar, with a number of key factors increasing the complexity of data collection. One of the assumptions made with respect to SAS image reconstruction is the presence of a constant sound speed. As a nearfield imaging system, SAS is sensitive to the breaking of this assumption. The sound speed in the ocean varies with depth. Variations in sound speed can come in the form of internal waves. Internal waves propagating up the slope of the continental shelf are subject to breaking mechanisms that result in the propagation of boluses shoreward. Internal wave boluses are three dimensional features consisting of colder, higher density water. Since the internal wave boluses are composed of colder seawater, the speed of sound is different than in the surrounding environment. The change in sound speed changes the timing and phase of propagating acoustic rays causing degradation in SAS image quality. Not only do the internal waves violate the constant sound speed assumption made by SAS for image formation, but they also influence the travel of acoustic rays due to a geometric lensing effect. The lensing effect causes large refractive effects near the top of the bolus, resulting in a bright region and shadow region within the image. The goal of this study was to quantify the effects of internal waves on SAS image resolution and subsequently model these effects. The quantification of the effects was performed utilizing point targets within the SAS image. The point spread function of the point targets was estimated and used as a proxy for the image resolution and showed that internal waves can cause resolution loss on the order of two to four times than in the absence of a bolus or sound speed error. A numerical ray tracing model was used to estimate the resolution loss in SAS imagery in the presence of internal waves. An analytical model derived in order to better characterize the impacts of internal waves on SAS resolution. Beamforming was also performed over simulated imagery in the presence and absence of internal waves. The models agreed well with each other and the observed resolution loss in collected SAS data. Based on the success of modeling attempts, it is reasonable to develop a method for full inversion for bolus parameters. Given the agreement of the models with data it may be possible to develop methods to compensate for timing errors caused by the presence of internal waves and return the ideal image resolution.

Download

Share

COinS