Funded by an Oak Ridge Associated Universities Powe Award
Dates: 2001-2002
Background/Research Objectives
In the ocean environment, sound provides a useful means of communication, surveillance, and scientific study. For example, an interesting scientific application of sonar is acoustic thermometry. The goal of thermometry is to infer changes in water temperature from changes in acoustic travel time (sound travels faster in warmer water). If acoustic signals are transmitted across an ocean basin, the result is an estimate of mean temperature that can be used to study climate changes such as global warming. One of the difficulties in implementing an acoustic climate monitoring system is that the ocean is not a static medium. Dynamic ocean processes such as internal waves produce temperature variations that translate into sound speed variations. Although these sound speed variations are small, they can have a profound effect on signals propagating over long distances.
Another sonar application that is strongly affected by fluctuations in the ocean environment is surveillance. Over the past 20 years substantial research has shown that incorporating a physics-based model of acoustic propagation into sonar systems can dramatically improve detection and localization capabilities. Performance degrades rapidly, however, when the model does not adequately capture the dynamic nature of the environment.
The key to improving sonar system performance in applications such as thermometry and surveillance is a thorough understanding of dynamic ocean processes. A stochastic model which relates fluctuations in the environment to fluctuations in acoustic signals is needed. Development of such a model is an interdisciplinary endeavor, requiring knowledge of physical oceanography, acoustics, and signal processing.
Motivation for my research comes from recent experience with the Acoustic Thermometry of Ocean Climate (ATOC) experiment. In ATOC a team of researchers transmitted sound from a low-frequency source to receiving arrays several thousand kilometers away. This experiment provided the first opportunity to observe broadband signals at megameter ranges for an extended period of time (almost a year). Using receptions at a range of 3515 km, I analyzed the signals propagating in the lowest modes of the waveguide, which constitute the most energetic arrivals at long ranges. That research produced the first detailed analysis of the variability of mode signals on short time and frequency scales. My ultimate goal is to develop a stochastic model for acoustic mode propagation through internal wave fields. As a step towards this goal, I propose to develop a method for estimating internal wave strength from measurements of the low-order modes at long ranges.