Present address: Great Lakes Institute for Environmental Research and Department of Biological Sciences, University of Windsor, 401 Sunset Avenue, Windsor, ON, Canada N9B 3P4.
Census vs. effective population size in chinook salmon: large- and small-scale environmental perturbation effects
Article first published online: 27 AUG 2003
Volume 12, Issue 10, pages 2571–2583, October 2003
How to Cite
Shrimpton, J. M. and Heath, D. D. (2003), Census vs. effective population size in chinook salmon: large- and small-scale environmental perturbation effects. Molecular Ecology, 12: 2571–2583. doi: 10.1046/j.1365-294X.2003.01932.x
- Issue published online: 5 SEP 2003
- Article first published online: 27 AUG 2003
- Received 28 November 2002; revision received 13 May 2003; accepted 11 June 2003
- census population size;
- chinook salmon;
- effective population size;
- genetic diversity;
Population viability has often been assessed by census of reproducing adults. Recently this method has been called into question and estimation of the effective population size (Ne) proposed as a complementary method to determine population health. We examined genetic diversity in five populations of chinook salmon (Oncorhynchus tshawytscha) from the upper Fraser River watershed (British Columbia, Canada) at 11 microsatellite loci over 20 years using DNA extracted from archived scale samples. We tested for changes in genetic diversity, calculated the ratio of the number of alleles to the range in allele size to give the statistic M, calculated Ne from the temporal change in allele frequency, used the maximum likelihood method to calculate effective population size (NeM), calculated the harmonic mean of population size, and compared these statistics to annual census estimates. Over the last two decades population size has increased in all five populations of chinook examined; however, Ne calculated for each population was low (81–691) and decreasing over the time interval measured. Values of NeM were low, but substantially higher than Ne calculated using the temporal method. The calculated values for M were generally low (M < 0.70), indicating recent population reductions for all five populations. Large-scale historic barriers to migration and development activities do not appear to account for the low values of Ne; however, available spawning area is positively correlated with Ne. Both Ne and M estimates indicate that these populations are potentially susceptible to inbreeding effects and may lack the ability to respond adaptively to stochastic events. Our findings question the practice of relying exclusively on census estimates for interpreting population health and show the importance of determining genetic diversity within populations.