Overview of Acoustic Methods and Technology, with Particular Reference to Groundfish Stocks

by Kenneth G. Foote Woods Hole Oceanographic Institution Woods Hole, MA 02543

Objectives of operational acoustic surveying

Five objectives of operational acoustic surveying are enumerated here, beginning with the elemental acoustic measurement and concluding with the search for associations of echo data with the bottom topography.

(1) Measurement of numerical density of fish concentration
The primary measurement is that of acoustic density in the sampled volume. The concentration of the target fish can be quantified, if sufficient information exists for separating the resulting numerical measures into those due to the target fish, i.e. signal, and those due to other scatterers, i.e. noise. Further knowledge is required to convert an acoustic value to a biological or physical measure of fish concentration.
(2) Description of geographic distribution
The geographic distribution of target fish or other organisms can be mapped through the basic acoustic measurements, as taxonomically classified.
(3) Abundance estimation
Estimates of abundance can be produced by integrating values of density over an area or throughout a volume, depending on whether the density is expressed per unit surface area or per unit volume, respectively. This is often expressed as the total number or mass of target fish, further distinguished by size or age group.
(4) Observation of behavior
In addition to the mentioned quantitative application of acoustics, observation of behavior is often a goal of surveying. Examples of behavior that can be quantified are position in the water column, diurnal vertical movements, and seasonal horizontal migration.
(5) Association of water-column distribution with bottom topography
In the case of groundfish especially, the defining behavioral orientation is to the bottom. Association of the distribution with bottom features, for example, depth, roughness, and particular features, can be very useful for distinguishing different kinds of groundfish when in a semipelagic state.

Basic acoustic methods

Transducer
The heart of any acoustic system is the transducer. By definition, this device converts one form of energy to another and vice versa. In the case of transducers used underwater, one of the most common transducing mechanisms is that of piezoelectricity. Accordingly, a pressure wave incident on the transducing material generates a voltage, and application of a voltage across the material generates a change in shape, launching an acoustic, or pressure, wave into the water.
Scientific echo sounder
Electronics are designed specifically to control the transmission and reception of pressure or acoustic waves in water. To launch an acoustic wave, transmitter electronics are designed to control the electrical excitation of the transducer in a very particular way. Similarly, electronics are designed to sense and register the voltage induced in a transducer by the incident acoustic wave. The combined package of transmitting and receiving electronics is called an echo sounder. When the various processes are sufficiently controlled or precise so that the echo sounder can be calibrated and the received signal can be processed quantitatively, the electronics package is called a scientific echo sounder. The calibration can be accomplished, for example, by means of a standard target of known acoustic target strength.
Commercial fishery echo sounder
While scientific echo sounders may be used on board commercial fishing vessels, and sometimes are, their expense for this purpose is often prohibitive. Echo sounders with much simpler assemblies of electronics are designed specifically for use on board fishing boats. Often, the sole output is a non-quantitative video signal that is displayed on a color screen. Such a commercial fishery echo sounder cannot be calibrated, hence cannot be used for ordinary scientific work.
Echogram
Data derived from an echo sounder are most simply visualized in the form of an echogram. This is a two-dimensional display in which the echo resulting from a single ping is displayed along a single line, with the signal amplitude or intensity indicated by the degree of darkening or color-coding at a displacement corresponding to the respective echo range. Echoes due to successive pings are aligned. With vertically oriented, hull-mounted transducers, the echo range corresponds to depth.
Echo counting
Echoes can be counted when the targets are sufficiently dispersed in space. By dividing the number of echo counts in a given region of the echogram by the sampling volume, a measure of target density is derived.
Echo integration
For the general situation of arbitrary density, as with a school of fish, individual target echoes may not be resolvable. In this case, the echo voltage, after detection and application of suitable range compensation, is squared and summed over a defined interval. The resulting quantity is proportional to the area backscattering coefficient. Division of this by the characteristic backscattering cross section for the target fish yields the numerical density.
Postprocessing echo sounder data
Acoustic data are generally voluminous. Echo data collected with a narrowband echo sounder at a single frequency with 0.1-m resolution over a 500-m range interval at 1 ping/s will generate data at the rate of 72 Mbyte/hour, assuming that each echo datum is represented by two bytes. Processing such data can be overwhelming without the aid of postprocessing software. This is generally available in the form of commercial software packages. It is remarked that all important decisions regarding the quality and allocation of the echo record are taken by the operator in an interactive mode. Typical operations performed with the postprocessing system include correcting for false bottom detections, assigning echo features to scatterer classes, e.g., targets and non-targets, recomputation of values of the area backscattering coefficient, and storing values of the coefficient in a database with high resolution in depth and sailed distance.
Other sonar types
In addition to echo sounders, other sonars are in use that could be adapted to quantitative fish-surveying. These sonar types include mechanically scanned sonar, sidescan sonar, interferometric sidescan sonar, multibeam sonar, Doppler sonar, acoustic Doppler profiler (ADP), and acoustic Doppler velocity profiler (ADVP).

Key issues

The two key issues in the acoustic quantification of fish are target strength and fish behavior. Target strength is defined as a logarithmic measure of backscattering cross section. Target strength depends on species, size, shape, biological condition, orientation, and sonar frequency. Its in-situ measurement and modeling present constant challenges to the researcher. Fish behavior, or how the fish is moving and orienting itself, or reacting to the passage of the transducer-bearing platform, e.g., research vessel, is also of concern in acoustic surveys.

Limitations

There are three major limitations in the acoustic surveying of fish, beyond those of knowing target strength and fish behavior.

(1) Registration of fish near the bottom
As soon as an echo is received from the bottom, all other echoes are lost, that is, become indistinguishable from the relatively enormous bottom echo. Fish that live on, or so close to, the bottom that the echo from bottom features returns to the transducer before the fish echoes do, will escape detection. The near-bottom zone where fish cannot be detected is popularly termed the acoustic "dead zone."
(2) Classifying fish echoes with narrowband sonar
The most common form of echo sounder uses a resonant transducer. This can only transmit signals of quite narrow bandwidth and receive echoes with strong filtering of the spectral content away from the resonance frequency. It is generally impossible to classify single echoes for want of bandwidth. Use of multiple echoes, information on proximity of fish echoes to the sea surface or bottom, and bandwidth can aid classification.
(3) Resolving individual scatterers except in a dispersed state
Direct determination of target strength requires resolution of individual fish echoes. This condition is often unobservable, owing to the numerical density of fish aggregation, its range from the transducer, and/or the transducer beamwidth.

Recent developments

A number of recent developments are addressing outstanding issues. These include special studies of target strength and fish behavior, collection of data at multiple frequencies, removal of noise from echograms, observation of fish near to the bottom with a towed transducer, application of multibeam sonar to fish quantification, and bottom-habitat characterization. Advances in multibeam sonar calibration are rendering this sonar useable in fish surveying.

National resources

A number of resources exist that may be drawn upon in developing acoustic methods for surveying groundfish. These include on-going acoustic survey programs at NMFS, research vessels carrying scientific echo sounders, echo-data postprocessing systems that are in routine use, fishing vessels available as platforms of opportunity, research programs engaging fishermen, acoustic calibration techniques and facilities, research and operational expertise, sonar manufacturing businesses, and the Alliance for Coastal Technologies, with its organizational and electronic communication resources.

Acknowledgments

We would like to thank K. Peterson and H. Sosik for images and comments.

The NOAA Alliance for Coastal Technologies (ACT) is thanked for their sponsorship of a workshop in February 2003, hosted by Gulf of Maine Ocean Observing System (GoMOOS), where this material was first presented. This report has been excerpted from the original presentation entitled "Overview of acoustic methods." [author Kenneth G. Foote, Woods Hole Oceanographic Institution, Woods Hole, MA].

Related Links

http://www.act-us.info/workshops_reports.php
[see GoMOOS February 26-28 2003]