Advancedpopulations

Metapopulations: Persistence Through Turnover

Overview

Many species live not as one continuous population but as metapopulations — networks of local populations in habitat patches, linked by dispersal, where local extinctions are balanced by recolonizations. Richard Levins formalized this in 1969: the species persists regionally even though no single patch population is permanent. The Glanville fritillary butterfly in Finland's Åland Islands — thousands of meadow patches winking in and out of occupancy — is the textbook example. Metapopulation theory transformed conservation: it explains why connectivity matters, predicts extinction thresholds from habitat loss, and guides reserve network design.

Learning objectives

  • Define metapopulation and its patch-occupancy dynamics
  • Explain Levins' model and the extinction–colonization balance
  • Distinguish classic, mainland–island, and source–sink metapopulations
  • Explain the extinction threshold from habitat destruction
  • Apply metapopulation theory to corridor design and reintroductions

Populations of Populations

A metapopulation is a set of local populations occupying discrete habitat patches, connected by dispersing individuals. Each patch population may go extinct — from demographic stochasticity, inbreeding, or local catastrophe — but dispersers from occupied patches recolonize empty ones. Regional persistence requires colonization to keep pace with extinction. This reframes conservation: saving a species means maintaining the network's dynamics, not just protecting individual patches. The Åland fritillary system, monitored across 4,000+ patches for decades, shows the theory in action.

Levins' rule

In the simplest model, the fraction of occupied patches equilibrates at p* = 1 − e/c, where e is extinction rate and c colonization rate. Persistence requires colonization to exceed extinction (c > e).

Three Metapopulation Structures

The classic (Levins) metapopulation has roughly equal patches with frequent turnover. Mainland–island systems have one large permanent source population supplying colonists to small satellite patches that blink in and out. Source–sink systems differ subtly: sinks are patches where deaths exceed births, persisting only through immigration — they can drain the metapopulation if they attract dispersers away from productive habitat (ecological traps). Identifying which structure applies determines whether to prioritize big sources or network connectivity.

The Extinction Threshold

Levins' model yields a stark prediction: if habitat destruction reduces patch availability below e/c, colonization cannot compensate and the metapopulation collapses — even though suitable patches remain. This extinction threshold means species can be doomed by habitat loss well before the last patch disappears, and the decline lags behind destruction (extinction debt again). It is a quantitative argument for protecting habitat networks above critical thresholds, not just remnant scraps.

What Makes a Good Patch Network

Persistence improves with: more patches (redundancy), larger patches (lower extinction), closer patches (higher colonization), and habitat quality (higher local growth). Corridors and stepping-stone patches boost colonization; matrix quality matters — dispersers crossing hostile farmland die, while those crossing semi-natural habitat survive. Hanski's incidence function model, fitted to real patch networks, predicts occupancy from patch area and isolation — a practical tool for ranking which patches to protect.

Conservation Applications

Metapopulation thinking guides: reserve networks designed for connectivity rather than isolated jewels; reintroductions that establish multiple linked populations instead of one vulnerable stock; managed relocation creating new nodes; and invasive-species control that accounts for recolonization (removing rats from one island fails if neighbors resupply them). It also warns against counting occupied patches as success — a metapopulation with declining colonization is heading for the threshold even while patches remain occupied.

Key points

  • Metapopulations persist regionally through local extinction–recolonization balance.
  • Levins' equilibrium: p* = 1 − e/c; persistence needs colonization > extinction.
  • Classic, mainland–island, and source–sink structures need different strategies.
  • Habitat loss below the extinction threshold collapses the network despite remaining patches.
  • Connectivity, patch size, and matrix quality jointly determine persistence.
  • Guides reserve networks, reintroductions, and invasive eradications.

Common misconceptions

✗ Myth: Protecting the currently occupied patches is enough.

✓ Truth: Empty patches are essential — they are future habitat for recolonization. A metapopulation needs the full network including vacant patches; protecting only occupied ones guarantees decline as extinctions go unreplaced.

✗ Myth: Local extinction means conservation failed.

✓ Truth: Local extinctions are normal metapopulation dynamics. Failure is when colonization stops compensating — the network-level balance is what matters, not any single patch.

✗ Myth: Source–sink and mainland–island metapopulations are the same.

✓ Truth: Mainland–island has a permanent large source; source–sink patches differ in demographic quality (sinks need immigration to persist). Sinks can even harm the metapopulation as ecological traps — a crucial distinction for management.

Practice questions

Q1. In Levins' metapopulation model, the species persists regionally when:
  1. Every patch is permanently occupied
  2. Colonization rate exceeds extinction rate (c > e)
  3. There is only one large patch
  4. Dispersal is zero
Answer: B.

Equilibrium occupancy p* = 1 − e/c is positive only if c > e. Local extinctions are expected and normal; regional persistence comes from recolonization outpacing extinction across the network.

Q2. The extinction threshold in metapopulation theory means:
  1. Species go extinct instantly
  2. Below a critical fraction of habitat, the metapopulation collapses even though patches remain
  3. Only predators cause extinction
  4. Thresholds don't apply to insects
Answer: B.

When habitat destruction pushes available patches below e/c, colonization cannot balance extinction and p* hits zero. Species can be doomed while suitable patches still exist — with the collapse lagging as extinction debt.

Q3. In a source–sink metapopulation, sink patches:
  1. Produce surplus dispersers
  2. Have deaths exceeding births and persist only via immigration
  3. Are always the largest patches
  4. Never affect the metapopulation
Answer: B.

Sinks are demographic deficits sustained by immigrants from sources. They can become ecological traps if attractive, draining the metapopulation — so identifying and managing sinks matters for conservation.

Q4. The Glanville fritillary study in Åland is important because it:
  1. Saved the species from extinction
  2. Provided decades of patch-occupancy data confirming metapopulation dynamics empirically
  3. Disproved Levins' model
  4. Studied only one meadow
Answer: B.

Monitoring 4,000+ patches over decades revealed ~500 occupied with constant turnover — local extinctions balanced by colonizations, exactly the dynamic equilibrium Levins predicted, making it the definitive empirical metapopulation.

Click a question to reveal the answer and explanation.

Frequently asked questions

What is the difference between a metapopulation and a subdivided population?

A metapopulation has discrete local populations with meaningful extinction–recolonization dynamics and limited dispersal between patches. A merely subdivided population has continuous distribution with local density variation but no patch-level turnover. The distinction matters because only metapopulations face network-level extinction thresholds.

What is an ecological trap?

An ecological trap is poor-quality habitat that organisms preferentially settle in, drawn by misleading cues — e.g. birds nesting in attractive but predator-rich suburbs. As sink habitat, traps drain dispersers from productive areas and can depress the whole metapopulation.

How does the rescue effect stabilize metapopulations?

Immigration rescues declining patch populations before they go extinct — boosting numbers, restoring genetic diversity, and recolonizing after crashes. Connectivity thus reduces effective extinction rates, which is why corridors strengthen metapopulation persistence beyond just enabling recolonization.

Can metapopulation theory guide reintroductions?

Yes: establish multiple linked populations rather than one, choose well-connected sites with good matrix, ensure patch sizes above viability thresholds, and plan for natural turnover — monitoring colonization rates, not just initial survival, as the success metric.