Tiny estimates of the Ne/N ratio in marine fishes: Are they real?

• Published in: Journal of fish biology Vol. 89; no. 6; pp. 2479 - 2504
• Main Author: Waples, R. S.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01.12.2016; Wiley Subscription Services, Inc
• Subjects: age structure; Animal reproduction; effective population size; Estimates; Fecundity; Fish; Fish populations; Hypotheses; Marine fish; overlapping generations; Ratios; Reproduction; sweepstakes reproductive success
• ISSN: 0022-1112; 1095-8649; 1095-8649
• DOI: 10.1111/jfb.13143

Theory and empirical estimates agree that the ratio of effective size (Ne) to census size (N) falls roughly in the range 0·1–0·5 for most populations. In a number of marine species, however, genetic estimates of contemporary Ne/N are as much as 5–6 orders of magnitude lower. Although some mechanisms that could produce such tiny Ne/N ratios have been proposed, the subject remains controversial. This issue is important to resolve: if Ne/N can be 10−3 or smaller, marine fish populations that are quite large could be at genetic risk. Based on a recently‐improved understanding of factors that influence Ne and Ne/N in species with overlapping generations, this paper evaluates conditions necessary to produce tiny Ne/N ratios in actual populations. These analyses show that although increased longevity, fecundity and variance in reproductive success that increase with age, and increased egg quality with age [the big old fat fecund female fish (BOFFFF) hypothesis] all reduce Ne/N, extreme scenarios are required to reduce Ne/N below about 0·01. Therefore, tiny Ne/N ratios require some version of Hedgecock's ‘sweepstakes’ hypothesis, whereby only a few families reproduce successfully. Simulations using common genetically‐based estimators show that, when true Ne is very large (≥106), a substantial fraction of point estimates of Ne/N can be 10−3 or smaller. These results mean that tiny, genetically‐based point estimates of Ne/N in large marine populations are expected to be quite common, even when the true Ne/N ratio is ‘normal’ (∼0·1 or higher). Very large samples of individuals can reduce, but not eliminate, this problem. The simulation results also emphasize the importance of considering deviations from model assumptions (e.g. non‐random sampling; weak selection or migration) that may be relatively small (and hence can generally be ignored when the signal is strong) but can lead to substantial biases when the drift signal is weak, as is likely for large marine populations. Empirical studies of this topic need to be able to distinguish between episodes of sweepstakes reproductive success that are ephemeral and lead to chaotic genetic patchiness, and those that are consistent enough across space and time to produce persistent evolutionary consequences.