| Summary: | Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2008. === Includes bibliographical references (leaves 221-229). === The application of a technique to instantaneously image and continuously monitor the abundance, spatial distribution, and behavior of fish populations over thousands of square kilometers using Ocean Acoustic Waveguide Remote Sensing (OAWRS) is demonstrated with data from its first implementation in a 2003 field experiment off the US Continental Shelf south of Long Island, NY. Conventional methods for monitoring fish populations rely on highly-localized, point measurements made from slow-moving research vessels that survey along widely spaced line transects to cover the vast ocean environments that fish inhabit and so significantly under-sample fish populations in time and space. This leads to incomplete, ambiguous and highly-aliased records of fish abundance and behavior. In contrast, OAWRS surveys at a rate roughly one million times greater than that of conventional fish-finding methods. Within a minute and a half, OAWRS images the ocean environment over more than ten thousand square kilometers, an area similar to the state of Massachusetts. This is possible because OAWRS exploits the natural capacity of the continental-shelf environment to act as a waveguide where sound waves are efficiently propagated over long ranges (tens of kilometers) via trapped modes that suffer only cylindrical spreading loss rather than the spherical spreading loss suffered in the short-range (hundreds of meters), waterborne propagation paths employed by conventional fish-finding sonar (CFFS). In this thesis, a method is developed for estimating the instantaneous population density and abundance of fish populations from wide-area OAWRS imagery. === The OAWRS population density estimates are calibrated with simultaneous local CFFS measurements, and are used to estimate the expected scattering cross section of an individual fish at OAWRS frequencies so that population density may be estimated in regions where CFFS measurements were not made. It is shown that the OAWRS population density estimates have uncertainties of less than 25% at each pixel or spatial resolution cell, for statistically stationary populations. Instantaneous abundance estimates then have much lower uncertainties when OAWRS population density is integrated over tens to hundreds of independent spatial resolution cells by the law of large numbers. A number of discoveries are also documented about the instantaneous horizontal structural characteristics, temporal evolution, short-term volatile behavior and propagation of information in very large fish shoals containing tens of millions of fish and spanning several kilometers in spatial extent. The OAWRS approach should enable new abilities in the study and assessment of fish populations and their behavioral dynamics. === by Deanelle T. Symonds. === Ph.D.
|
|---|
