Abstract
This paper applies BASCET, a Bayesian Spatial Composition Estimation Tool for clusters of acoustically identified schools, to Bering Sea acoustic survey data collected during 1994. As the method employs prior information from an acoustic expert, procedures for eliciting such information are suggested and pitfalls of the process are indicated. Techniques for model checking using the posterior predictive distribution are employed, as is a multi-chain method for evaluating the convergence of the Markov-Chain Monte Carlo algorithm used in BASCET. Unlike methods based on neural networks, BASCET is able to provide confidence regions for its estimates of school cluster composition. In addition, it can indicate which school cluster attributes were most influential in determining a given estimate, a useful tool for model checking that is here demonstrated on a randomly selected cluster. Estimated abundance ratios of juvenile to adult pollock (Theragra chalcogramma) were compared, in two regions, to the values used by expert technicians. Ratios differed from expert values by less than 0.03 in both regions. The encouraging results reported here suggest that the BASCET method, originally tested on simulated data, may be usefully applied to real surveys.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have