Publication Abstract

A simple statistical method for catch comparison studies

Rene Holst and Andy Revill*

This study reports on the first known testing in European waters of a new Rhode Island (USA) design of fishing trawl, known as the ‘Eliminator’. We found that this trawl could be used to selectively harvest haddock (Melanogrammus aeglefinus) and whiting (Merlangius merlangus) in a mixed-demersal North Sea fishery while allowing cod (Gadus morhua) and other species to escape. The number of cod caught were reduced by 89% and accounted for 2% (weight) of the marketable fish caught in the Eliminator, contrasted by 10 % weight in the control trawl. We note that North Sea stocks of cod are overexploited and at risk of being fished unsustainably, whereas haddock stocks have full reproductive capacity with a biomass above precautionary limits. We suggest that this trawl may have potential utility within the ongoing Northern European cod recovery programme.

For the analysis of our catch comparison data from these trials, we propose a simple method based on Generalised Linear Mixed Models (GLMM) and use polynomial approximations to fit the proportions caught in the test codend. The method provides comparisons of fish catch at length by the two gears through a continuous curve with a realistic confidence band. We demonstrate the versatility of this method, as our data includes a range of species with different selective patterns.

Reference

Rene Holst and Andy Revill* (2009) A simple statistical method for catch comparison studies. Fisheries Research 95 (2-3) 254-259