© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Analysis of non-linear relationships between catch per unit effort and abundance in a tuna purse-seine fishery simulated with artificial neural networks
a Institut de Recherche pour le Développement (IRD) UR 109 CRHMT BP 171, 34203 Sete Cedex, France
b Instituto Nacional de la Pesca-PNAAPD, Facultad de Ciencias Marinas UABC, Mexico
*Correspondence to D. Gaertner: tel.: +33 499 573231; fax: +33 499 573295. e-mail: gaertner{at}ird.fr.
A simulation study, combining grid- and individual-based approaches, was conducted to analyse the shape of the relationship between catch per unit effort (cpue) and abundance in a tuna purse-seine fishery. To understand the effect of fleet dynamics on the interpretation of cpue, the decision-making process used by fishers while searching for the resource is modelled with artificial neural networks. The cpue of fishers operating independently (i.e. individuals) vs. fishers sharing information (i.e. a code-group) is compared, accounting for different environmental scenarios. The results show that a power curve non-proportional relationship between cpue and abundance performs better than a linear relationship. As the shape parameter of the power curve for the code-group fishers was lower in every scenario than that of individual fishers, we conclude that hyperstability, a phenomenon commonly observed in schooling fisheries, is mainly attributable to information exchange among vessels. Setting the individual-level state variables of the virtual system at a specific spatial and temporal scale may affect the results of the simulations.
Keywords: artificial neural networks, cpue–abundance relationship, fisher behaviour, information-sharing, tuna fishery
Received 25 November 2003; accepted 28 April 2004.