A dynamic model of the Bay of Biscay pelagic fleet simulating fishing trip choice: the response to the closure of the European anchovy (Engraulis encrasicolus) fishery in 2005

Abstract

International audienceThe scope of this paper is to describe, evaluate, and forecast fishing trip choices of the Bay of Biscay pelagic fleet using random utility modeling (RUM). First, alternative fishing trip choices of this fleet were identified using multivariate statistical methods based on species landings weighted by value and defined as distinct fishing activity or fisheries (termed métiers). A RUM was specified, which included trip components as attributes during the period 2001-2004 (a lagged percentage of the value per unit of effort of the main species caught, total value per unit of effort, and inertia in terms of changes from one métier to another). For the main métiers, the proportion of correct effort allocation is 90% during the calibration period of 2001-2004. The results from the RUM are used to parameterize a simulation model of trip choice. The model is used to predict trip choices in 2005, throughout most of which fishing was constrained by the closure of the European anchovy (Engraulis encrasicolus) fishery. Simulation results are compared with observed trip choices following the fishing ban: 80% of observed trip choices are correctly predicted by the model. The capacity of the behavioral model to predict responses to the closure is then discussed.Cet article a pour objectif de décrire, d'évaluer et de prédire le choix de l'activité de pêche à l'échelle de la marée pour les chalutiers pélagiques du Golfe de Gascogne en utilisant des modèles de choix discrets (RUM, « random utility modeling »). Les pratiques alternatives ont été identifiées par des méthodes statistiques multivariées basées sur les débarquements spécifiques en valeur et ont été définies comme autant de pratiques nommées par la suite métiers. Un RUM a été spécifié incluant les caractéristiques de la marée (le pourcentage de la valeur par unité d'effort (VPUE) des espèces principales capturées et la VPUE totale de la marée précédente ainsi que l'inertie pour changer de métier) comme attributs durant la période 2001-2004. Pour les métiers principaux, le modèle est capable de prédire correctement 90 % des choix observés durant la période de calibration. Nous utilisons ensuite ces résultats pour paramètrer un modèle de simulation de choix d'activité. Le modèle a servi à prédire les choix en 2005, année au cours de laquelle un arrêt de pêche à l'anchois (Engraulis encrasicolus) a été imposé. Les résultats de la simulation sont comparés aux choix observés suite à la fermeture : 80 % des choix de métiers observés sont correctement prédits par le modèle. La capacité du modèle à prédire la réponse à la fermeture fait ensuite l'objet d'une discussion

Fishing fleet typology, economic dependence, and species landing profiles of the French fleets in the Bay of Biscay, 2000-2006

“…Offshore pelagic trawlers switched from anchovy to albacore and to a lesser degree to sea bass and mackerel when the anchovy fishery was closed in 2005. These results confirm the trip-level analysis of the same pelagic trawlers and some purse seiners carried out by Vermard et al (2008), who analysed fishing trip choices during the period 2000-2005. The choice concerned the main target species of each trip, leading to the identification of four métiers defined by anchovy, see bass, albacore and a mixture of species.…”

Predicting fisher response to competition for space and resources in a mixed demersal fishery

“…In the present case, two distributions have been considered. First, a conditional logit model (McFadden, 1974;Vermard et al, 2008) was used. This is the simplest sort of distributions to be considered, and also the one for which model outcomes are the easiest to interpret.…”

“…RUMs provide an appropriate and functional approach to describe how fishers make a choice among a panel of finite alternatives (Wilen et al, 2002). Such a discrete-choice modelling approach has been applied to analyse fishers' choice of fishing ground (Hutton et al, 2004;Wilen et al, 2002), target species (Pradhan and Leung, 2004a;Vermard et al, 2008), and gear type (Andersen et al, 2012;Holland and Sutinen, 1999;Marchal et al, 2009). …”

Modelling climate-change effects on Australian and Pacific aquatic ecosystems: a review of analytical tools and management implications

“…Moreover, quantitative models of fisher behaviour and fishing fleets are available for a range of fisheries around the world (e.g. Vermard et al 2008). Using statistical or Markov chain fleet-level approaches have made models more realistic because they rely less on historical information that could be less applicable in non-linear climatechange scenarios.…”