Data from: Spatiotemporal relationship between adult census size and genetic population size across a wide population size gradient

• Main Authors: Bernos, Thaïs A., Fraser, Dylan J.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: 2016
• Subjects: Effective number of breeders; census population size; Stream fish; mark-recapture; genetic monitoring; Effective population size; Population Genetics - Empirical; Population Ecology; Conservation Genetics; Canada; Newfoundland
• Online Access: http://hdl.handle.net/10255/dryad.121533; https://doi.org/10.5061/dryad.f9f75

Adult census population size (N) and effective number of breeders (Nb) are highly relevant for designing effective conservation strategies. Both parameters are often challenging to quantify, however, making it of interest to determine whether one parameter can be generalized from the other. Yet, the spatiotemporal relationship between N and Nb has not been well characterized empirically in many taxa. We analysed this relationship for 5–7 consecutive years in twelve brook trout populations varying greatly in N (49-10032) and Nb (3-567) and identified major environmental variables affecting the two parameters. N or habitat size alone explained 47–57% of the variance in Nb, and Nb was strongly correlated with effective population size. The ratio Nb/N ranged from 0.01 to 0.45 and increased at small N or following an annual decrease in N, suggesting density-dependent constraints on Nb. We found no evidence for a consistent, directional difference between variability in Nb and/or Nb/N among small and large populations; however, small populations had more varying temporal variability in Nb/N ratios than large populations. Finally, Nb and Nb/N were 2.5- and 2.3-fold more variable among populations than temporally within populations. Our results demonstrate a clear linkage between demographic and evolutionary parameters, suggesting that Nb could be used to approximate N (or vice versa) in natural populations. Nevertheless, using one variable to infer the other to monitor trends within populations is less recommended, perhaps even less so in small populations given their less predictable Nb vs. N dynamics.