Effective population size

Effective population size (N_e) is a concept in population genetics that measures the size of an idealized population that would experience the same magnitude of genetic drift as the actual population under study. The actual population may be larger than N_e due to factors that reduce genetic diversity: unequal sex ratios, variation in family size, population bottlenecks, and non-random mating.

The effective population size determines the balance between natural selection and genetic drift. When N_e is large, selection dominates and advantageous alleles spread reliably. When N_e is small, drift can overpower selection and drive alleles to fixation regardless of their fitness effects. This is why balancing selection can maintain polymorphisms in large populations but may fail in small ones.

The concept was developed by Sewall Wright and remains central to understanding how population structure shapes evolutionary outcomes. In conservation biology, N_e is the key parameter for predicting the viability of endangered populations; in evolutionary theory, it sets the threshold for when selection becomes effective.

Why N_e Differs from Census Size

A population's census size (N) — the actual number of breeding individuals — is almost always larger than its effective size. The ratio N_e/N is typically between 0.1 and 0.5 in natural populations, and it can be much lower in species with extreme mating systems or highly variable reproductive success.

The factors that reduce N_e relative to N include:

Unequal sex ratios: When one sex is rare, the effective size is determined by the rarer sex. A population with 100 males and 10 females has an effective size closer to 40 than to 110, because the females are the reproductive bottleneck.

Variation in family size: In populations where a few individuals produce many offspring and most produce few, the effective size is reduced because the genetic contributions are concentrated in a small number of lineages. This is common in species with high fecundity and high juvenile mortality.

Population bottlenecks: When a population passes through a period of very small size — due to disease, habitat loss, or founder events — its effective size is permanently reduced, even if the population subsequently recovers to a large census size. The genetic diversity lost during the bottleneck is not regained by population growth.

Non-random mating: Inbreeding and population subdivision both reduce effective size. Inbreeding increases homozygosity and accelerates the loss of rare alleles. Population subdivision isolates gene pools and prevents the spread of beneficial alleles across the species range.

Overlapping generations: When generations overlap, the effective size is a harmonic mean of the effective sizes across age classes, weighted by their reproductive value. This means that a population with many old, non-reproducing individuals and few young breeders has a lower effective size than its census count would suggest.

The Threshold for Selection

The most consequential property of effective population size is that it sets a threshold for the efficacy of natural selection. In a large population, a new beneficial mutation with selection coefficient s spreads with high probability if 4N_e s >> 1. In a small population, the same mutation is likely to be lost by drift unless 4N_e s is large.

This threshold has profound implications. It means that mildly beneficial mutations — those with small s — are effectively neutral in small populations. They drift to fixation or loss without selection having any influence. Only strongly beneficial mutations or those that arise at high frequency can escape drift in small populations. The result is that small populations adapt primarily through rare, large-effect mutations, while large populations can accumulate adaptation through many small-effect mutations.

The threshold also explains why genetic load accumulates in small populations. Deleterious mutations with selection coefficients smaller than approximately 1/(2N_e) behave as effectively neutral and drift to fixation. A population with N_e = 50 will fix deleterious mutations with s < 0.01, producing a fitness decline that accelerates as N_e shrinks further. This is the mechanism of mutational meltdown: small populations lose fitness through the accumulation of mildly deleterious alleles, which reduces N_e further, which accelerates the accumulation of deleterious alleles, in a positive feedback loop toward extinction.

Measuring N_e

Effective population size can be estimated through several methods, each capturing a different aspect of the concept:

Genetic methods use temporal changes in allele frequencies, heterozygosity, or linkage disequilibrium to infer N_e. The heterozygote-excess method, the linkage-disequilibrium method, and the temporal-method-of-moments all produce estimates that reflect the recent effective size — the size over the past few to tens of generations. These methods are widely used in conservation genetics because they require only genetic samples, not demographic data.

Demographic methods calculate N_e from life-table data: sex ratios, variance in reproductive success, age structure, and population size fluctuations. The demographic estimate reflects the theoretical effective size given the observed mating system and population structure. It may differ from the genetic estimate if the population has experienced recent bottlenecks or gene flow that are not captured in the current demography.

Coalescent methods use phylogenetic or population-genomic data to estimate N_e over deeper timescales. These methods model the ancestral process backward in time and infer the effective size history that best explains the observed patterns of genetic variation. They are particularly useful for detecting historical bottlenecks or expansions that demographic censuses cannot reveal.

The different methods often produce different estimates, and the discrepancy is informative. If genetic N_e is much smaller than demographic N_e, the population has likely experienced a recent bottleneck or severe family-size variance. If genetic N_e is larger, there may be gene flow from unsampled populations or historical population structure that inflates genetic diversity.

N_e in Conservation and Management

Conservation biology operates under a rule of thumb: maintain N_e above 50 to avoid inbreeding depression, and above 500 to preserve long-term evolutionary potential. These numbers — derived from theoretical models and empirical observation — are not arbitrary thresholds but approximate boundaries where drift begins to dominate selection and where genetic load becomes unsustainable.

The 50/500 rule has been criticized as overly simplistic. Real populations are structured, fluctuate in size, experience gene flow, and have complex mating systems that the simple model does not capture. But the rule persists because it captures a genuine qualitative transition: below N_e ≈ 50, populations enter a regime where drift dominates and fitness declines accelerate; below N_e ≈ 500, the capacity for adaptive evolution is severely constrained.

In fisheries management, N_e is increasingly used to set catch limits. Overfishing reduces population size, which reduces N_e, which increases the fixation of deleterious alleles and reduces the population's capacity to adapt to changing ocean conditions. The collapse of the Newfoundland cod fishery in the 1990s was preceded by a severe reduction in effective population size, which left the stock genetically depleted and unable to recover even when fishing pressure was removed.

Effective population size is the lens through which the raw count of individuals becomes a meaningful evolutionary parameter. A population of ten thousand individuals with an N_e of fifty is not a large population. It is a small population wearing a large population's clothes — and the clothes do not fit.