An Evaluation of Methods for Indirectly Measuring the Mean Acoustic Scattering Cross Section of Fish

Acoustic techniques are being used throughout the world for assessment of fish stocks. The most common acoustic assessment technique is to process the output of an echo sounder using a procedure called echo integration. The echo integration technique is based on the fact that the integrated intensity of the acoustic signal scattered off a fish population is proportional to the product of the mean acoustic scattering cross section of the individual fish in the population and the number of fish acoustically illuminated. The mean acoustic scattering cross section must be known to obtain an absolute estimate of stock size using this technique. A great deal of effort has been expended on the development of methods for measuring this parameter. This paper discusses two methods for indirectly extracting the mean scattering cross section from the empirical distribution of single fish echo amplitudes. The first method discussed obtains the estimate of mean scattering cross section by first estimating the single fish target strength probability density function. The second method assumes that the square root of the scattering cross section is Rayleigh distributed. With this assumption, the estimation problem becomes one of determining the single parameter in the Rayleigh distribution. The performance of the two estimation techniques is evaluated using a Monte Carlo simulation.