Bimodal size distributions in Arctic char, Salvelinus alpinus : artefacts of biased sampling

• Published in: Canadian Journal of Fisheries and Aquatic Sciences
• Main Authors: Finstad, Anders Gravbrøt, Jansen, Peder Andreas, Hirvonen, Heikki
• Format: Article in Journal/Newspaper
• Language: English
• Published: Canadian Science Publishing 2003
• Subjects: Aquatic Science; Ecology, Evolution, Behavior and Systematics; Arctic; Salvelinus alpinus
• Online Access: http://dx.doi.org/10.1139/f03-089; http://www.nrcresearchpress.com/doi/pdf/10.1139/f03-089

Bimodal population size and age distributions in Arctic char (Salvelinus alpinus (L.)) and hypotheses on growth patterns generating bimodality have drawn considerable attention during the last decade. However, such bimodality has also been suggested to be an artefact of biased sampling. We examined published data sets reporting bimodal size distributions in gill-net samples of Arctic char in order to confront hypotheses on growth patterns generating bimodal population size distributions. Growth patterns were derived from published length-at-age data. Simulations revealed that the observed growth patterns evidently could not generate a bimodal population size distribution. The basic reason for this was that growth did not stagnate strongly enough in the largest size classes of Arctic char. The reliability of growth approximations from length-at-age data was supported by empirical data on back-calculated growth trajectories. Furthermore, differences in year-class strength cannot explain all of the observed bimodal size and age distributions in gill-net samples, as they have been reported to persist over time. Thus, bias in the sampling procedure, which overestimates the frequency of old and large fish, is retained as the only plausible explanation for stable bimodal size distributions often observed in Arctic char gill-net samples.