Fine-scale mapping of fish distribution from commercial catch and effort data using maximum entropy tomography

We present a technique for mapping the spatial distribution of fish using commercial catch and effort data. The relative fish density can be estimated at scales smaller than the length of the unit of effort, such as a tow, by using a Bayesian maximum entropy method. This can take advantage of the fact that the tows cross to give information about the density in those areas. This is a novel application of a well-tested technique that has been used in other fields such as astronomical imaging. Its utility and robustness is demonstrated both on simulated data and on real data from the trawl fishery on spawning hoki off the west coast of the South Island of New Zealand.