Key-words:catch-per-unit effort, length-frequency distribution, Procambarus clarkii, sampling method, trappingThe ecological importance of the red-swamp crayfish (Procambarus clarkii) in the functioning of freshwater aquatic ecosystems is becoming more evident. It is important to know the limitations of sampling methods targeting this species, because accurate determination of population characteristics is required for predicting the ecological success of P. clarkii and its potential impacts on invaded ecosystems. In the current study, we addressed the question of trap efficiency by comparing population structure provided by eight trap devices (varying in number and position of entrances, mesh size, trap size and construction materials) in three habitats (a pond, a reed bed and a grassland) in a French marsh in spring 2010. Based on a large collection of P. clarkii (n = 2091, 272 and 213 respectively in the pond, reed bed and grassland habitats), we found that semicylindrical traps made from 5.5 mm mesh galvanized steel wire (SCG) were the most efficient in terms of catch probability (96.7-100% compared to 15.7-82.8% depending on trap types and habitats) and catchper-unit effort (CPUE: 15.3, 6.0 and 5.1 crayfish·trap −1 ·24 h −1 compared to 0.2−4.4, 2.9 and 1.7 crayfish·trap −1 ·24 h −1 by the other types of fishing gear in the pond, reed bed and grassland respectively). The SCG trap was also the most effective for sampling all size classes, especially small individuals (carapace length 30 mm). Sex ratio was balanced in all cases. SCG could be considered as appropriate trapping gear to likely give more realistic information about P. clarkii population characteristics than many other trap types. Further investigation is needed to assess the catching effort required for ultimately proposing a standardised sampling method in a large range of habitats. −1 ·24 h −1 trouvées dans les autres pièges respectivement dans le plan d'eau, la roselière et la prairie inondée). Les pièges SCG étaient également les plus efficaces pour échantillonner toutes les classes de taille, et particulièrement les petites écrevisses (longueur de carapace 30 mm). Le sexe ratio était équilibré dans tous les cas. Les pièges SCG pourraient être considérés comme un engin de pêche approprié pour fournir des informations probablement plus réalistes sur les caractéristiques de populations de P. clarkii que de nombreux autres types de piège. Des études supplémentaires sont nécessaires afin d'évaluer l'effort d'échantillonnage requis pour finalement proposer une méthode d'échantillonnage standardisée dans une large gamme d'habitats.
RÉSUMÉ