Optimal swimming speeds and forward-assisted propulsion: energy-conserving behaviours of upriver-migrating adult salmon

Authors:
  • SHScott G. Hinch
    Scott G. Hinch
  • PRPeter S. Rand
    Peter S. Rand
  • KMKoji Miyoshi
    Koji Miyoshi
  • KHKazufumi Hayashida
    Kazufumi Hayashida
  • TSTaku Sakashita
    Taku Sakashita
  • MFMakoto Fujii
    Makoto Fujii
  • HNHisaya Nii
    Hisaya Nii
  • KNKatsuya Nakao
    Katsuya Nakao
  • HUHiroshi Ueda
    Hiroshi Ueda
  • TCTheodore Castro-Santos
    Theodore Castro-Santos
  • FSFrancisco Javier Sanz-Ronda
    Francisco Javier Sanz-Ronda
  • JRJorge Ruiz-Legazpi
    Jorge Ruiz-Legazpi

Abstract
Anadromous salmon migrations are energetically expensive. Long-distance migrants should be efficient in their use of energy and minimize swimming costs wherever possible. We explore swimming strategies and energy-saving tactics employed by three long-distance-migrating sockeye salmon (Oncorhynchus nerka) stocks in the Fraser River watershed, British Columbia. We used stereovideography and bank-side observations to estimate swimming speeds (from tailbeat frequency) and ground speeds (using distance traveled and duration) for individuals at several sites. Salmon were highly efficient at migration (i.e., ground speeds equaled or exceeded swimming speeds) through reaches with relatively low encountered currents (-1). We speculate that salmon exploit small reverse-flow vortices to achieve this feat. With low encountered currents, most salmon migrated according to an optimal swimming speed model: migrants minimized transport costs per unit distance traveled. Generally, salmon were less efficient at migration with fast currents, although the Chilko stock were superoptimal migrants, possibly owing to unique morphology and (or) behaviours. The risk of significant delays is enhanced when fast currents are encountered. Under these conditions, relatively fast swimming speeds could minimize travel time, despite high costs. Migrants may be balancing energetic costs of migration against the fitness costs of spawning delays.