An unbiased, stochastic echo-counting model
Traditional echo counting is an apparently simple fisheries acoustic tool that can provide quantitative estimates of fish density and abundance. The method has been widely used as it requires a minimum of equipment and the data analysis is deceptively simple.
Several echo-counting methods have been described in the literature. These generally assume that the fish are all of the same acoustic size. This unrealistic assumption has led us to develop a new. stochastic echo-counting model that accounts for the wide distribution in acoustic fish sizes that is typical even for a group of fish of the same size.
The new parametric model uses the Rayleigh or Rice probability density function to describe the wide distribution in acoustic fish size. Our major conclusions are that the traditional model may lead to a biased fish-density estimate and that better estimates of the mean acoustic fish size and other model parameters are very important.
