Migration Rate Calculator

The Migration Rate Calculator estimates how strongly gene flow connects two populations in population genetics. Migration, also called gene flow, is the movement of individuals (and the alleles they carry) from one population into another. This calculator quantifies that movement as a migration rate (m), the proportion of a recipient population that is made up of migrants each generation, and it links that rate to classic measures of population structure such as the number of effective migrants (Nm) and the expected fixation index (F_ST) at migration-drift equilibrium.

The tool offers two complementary methods. The Direct Count mode takes the raw number of migrants arriving per generation and the recipient population size, and returns the migration rate directly. The From Frequencies mode works backward from observed allele frequency change, inferring the migration rate that would shift a local allele frequency from an initial value toward the source population's frequency over a known number of generations. Both methods are grounded in Sewall Wright's island model of migration and are widely used in conservation genetics, evolutionary biology, and dispersal studies. Whether you are a student learning the Wright island model or a researcher screening candidate barriers to gene flow, this migration rate calculator turns counts and frequencies into interpretable gene flow statistics in seconds.

In Direct Count mode the migration rate is simply the fraction of the recipient population that consists of immigrants in a given generation. If you observe a number of migrants arriving and you know the size of the population receiving them, the migration rate m is the ratio of the two. The calculator then derives several downstream gene flow parameters from that rate.

The headline output is Nm, the product of population size and migration rate, which equals the absolute number of migrants per generation in the island model. Nm is the single most important number in this analysis because it governs whether gene flow or genetic drift dominates. The calculator also reports the expected F_ST at migration-drift equilibrium and an effective migration term m(1 - m) that accounts for the fact that arriving migrants dilute, but do not fully replace, the resident gene pool.

When you cannot count migrants directly, you can still infer the migration rate from how a local allele frequency drifts toward the frequency of a source population. Under the island model, each generation the local frequency moves a fraction m of the way toward the source frequency. After t generations the local frequency approaches the source frequency geometrically, and the calculator solves that geometric relationship for m.

The recursion is p′ = (1 - m)p + m×p_M, where p is the current local frequency and p_M is the source population frequency. Iterating this expression for t generations gives (1 - m)^t = (p₁ - p_M) / (p₀ - p_M), which the calculator rearranges to recover the per-generation migration rate. The equilibrium frequency that this process converges on is simply the source frequency p_M: as long as migration continues, the recipient population is eventually homogenized toward its source. The tool also reports a predicted final frequency by replaying the recursion, which should match your observed final frequency when the model fits.

The quantity Nm deserves special attention because it captures the tug-of-war between gene flow and genetic drift. A famous result from Sewall Wright, often called the one-migrant-per-generation rule, states that an Nm greater than one is usually enough to prevent populations from differentiating by drift alone. The migration rate calculator flags each result as High Gene Flow when Nm exceeds one and Low Gene Flow when it falls below one.

• Nm > 1: gene flow is strong enough to swamp drift, so populations remain genetically connected and F_ST stays low.
• Nm ≈ 1: migration and drift are roughly balanced, producing moderate differentiation.
• Nm < 1: drift dominates, allowing local populations to diverge and F_ST to climb.

It is important to remember that the one-migrant rule assumes an idealized island model with constant population sizes, selectively neutral alleles, and migration-drift equilibrium. Real systems deviate from these assumptions, so Nm should be read as a rough threshold rather than a hard cutoff. Even so, computing Nm with this calculator gives conservation biologists a fast, defensible way to gauge connectivity between habitat patches and to prioritize corridors that keep small populations from sliding into isolation.

Migration rate estimates underpin a wide range of biological questions. In conservation genetics, managers use gene flow estimates to decide whether fragmented populations are functioning as one connected unit or are drifting apart toward inbreeding and local extinction. A low migration rate between forest fragments, for example, can justify building wildlife corridors or undertaking assisted gene flow. In evolutionary biology, migration counteracts local adaptation: high gene flow can erode the genetic differences that natural selection builds between environments, while low gene flow lets populations specialize.

The frequency-based method is especially useful when working with historical or museum samples, where you have allele frequencies at two time points but no direct count of dispersers. The direct-count method suits field studies that mark and recapture individuals, count seeds or larvae arriving from upstream, or estimate immigrant numbers from telemetry. Both feed into the same downstream statistics, letting you compare gene flow across taxa with very different life histories. Population geneticists also pair these migration estimates with related tools, computing F_ST from sequence data, modeling effective population size, and projecting genetic drift, to build a complete picture of how dispersal, drift, and selection shape biodiversity over time.

Every output from this migration rate calculator rests on the assumptions of Wright's island model, and understanding those assumptions keeps your interpretation honest. The model assumes migration is symmetric and constant across generations, that the receiving population has a stable size, and that the alleles being tracked are selectively neutral. It also treats the census population size N as equal to the effective population size N_e, which is a simplification: real effective sizes are usually smaller than census counts because of unequal sex ratios, variance in reproductive success, and fluctuating population sizes.

The F_ST output is the value expected at equilibrium, not necessarily the value you would measure today; populations recovering from a recent bottleneck or a sudden change in connectivity may not have reached that balance yet. In frequency mode, the estimate also assumes the source frequency p_M is fixed and that drift and selection are negligible relative to migration over the interval studied. When the observed frequency overshoots the source frequency, or when initial and source frequencies are equal, the equation has no valid solution and the calculator returns no result. Treat the numbers as a transparent, model-based first approximation, and validate important conclusions with multilocus data and formal coalescent or Bayesian gene flow estimators.