Bimodal run distribution in a northern population of sockeye salmon (Oncorhynchus nerka): life history and genetic analysis on a temporal scale

Authors

  • E. K. Fillatre,

    1. Great Lakes Institute for Environmental Research and the, Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, Windsor, Ontario, Canada N9B 3P4,
    Search for more papers by this author
  • P. Etherton,

    1. Department of Fisheries and Oceans, Canada, Suite 100, 419 Range Road, Whitehorse, Yukon, Canada Y1A 3V1
    Search for more papers by this author
  • D. D. Heath

    Corresponding author
    1. Great Lakes Institute for Environmental Research and the, Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, Windsor, Ontario, Canada N9B 3P4,
    Search for more papers by this author

D. D. Heath. Fax: (519) 971 3616; E-mail: dheath@uwindsor.ca

Abstract

Life history variation and genetic differentiation were analysed in sockeye salmon in Klukshu River, Yukon Canada over 7 years (1994–2000). Sockeye salmon return to the Klukshu River in two distinct runs, with a small ‘early run’ in June–August, and a larger ‘late run’ in August–September. A maximum likelihood test for clusters indicated that the return frequency distribution was bimodal in all the years analysed. Life history differences (fork length, sex ratio, age at maturity, fresh- and saltwater residency times) were found between the early and late runs; however, inconsistent patterns suggest that environmental effects outweigh, or strongly interact with, genetic effects for the life history characters evaluated. Analysis of variation at eight microsatellite loci showed that the early and late runs are genetically differentiated in all years examined (exact test). FST estimates between runs within years were significantly greater than zero (range: 0.018–0.041) for all years except one (0.004). The genetic variance explained by early vs. late runs (2.27%) was twice the variance among years (1.16%) based on analysis of molecular variance. Our neighbour-joining tree showed early and late runs generally clustering separately, indicating higher gene flow among the early or late run fish across years relative to between-run gene flow. Two years did not fit the general clustering pattern; although the early and late runs in 1995 and 2000 were genetically differentiated, they clustered separately from the rest of the groups. We cannot offer a definitive explanation for these anomalies; however, an analysis of possible cryptic population structure in early and late runs indicated that at least a few fish strayed between the runs in each year, and the highest rate of mixing was in 1995 and 2000. Our data indicate that the runs are at least partially reproductively isolated as a result of temporal and/or spatial isolating mechanisms. Such reproductive isolation has important implications for conservation and management of the Klukshu sockeye salmon, and make them an evolutionarily interesting group because of parallels with incipient speciation.

Ancillary