# Island Biogeography

> An island's species count changes at the rate new species arrive minus the rate resident species are lost, so it settles near where the two rates are equal while the species themselves keep being replaced. This lesson builds that MacArthur-Wilson equilibrium from an immigration curve and an extinction curve, reads the Simberloff-Wilson mangrove defaunation data from the Florida Keys, and separates richness from identity with a Jaccard comparison. Afterwards you can find an equilibrium species number and its turnover from two rate curves, and say what the model can and cannot tell a planner about habitat fragments and corridors.

Course: Ecology and Evolution (Interactions and communities). Level: core university (third year). About 17 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/island-biogeography/

Builds on: Population Growth and Regulation · Distributions, Niches and Tolerances

## What you'll learn

- Explain the MacArthur-Wilson equilibrium: why immigration of new species falls and extinction rises as richness grows, where the equilibrium lies, and how distance and area shift it.
- Calculate an equilibrium species number, its turnover rate and the approach to equilibrium from stated immigration and extinction curves.
- Read the Simberloff-Wilson mangrove defaunation results and explain why returning richness with changed species lists and continuing turnover is consistent with a dynamic equilibrium.
- Distinguish richness from identity with a Jaccard similarity, and apply the model, with its limits, to habitat fragments and corridors.

The first film showed why near, large islands hold the most species, but not how fast an emptied island refills, or whether the returning species are the ones that left.

> **The one idea.** An island’s species count changes at the rate new species arrive minus the rate resident species are lost. Arrivals of new species slow as the island fills and losses speed up, so the count settles near where the two rates are equal, while the species themselves keep being replaced.
>
> *Scope:* one island fed by a mainland (or nearby islands), with new species forming on the island left out (Going deeper); this scope holds throughout. It is a model of rates: it predicts how many species and how fast they turn over, not which ones. Habitat fragments on land are only partly islands.

## Two rates and a crossing

A cistern with a float valve works the same way: as the water rises, the float throttles the inflow and the drain runs faster, so the level settles while the water keeps changing.

> **Equation card: How fast the number of species on an island is changing, from the two rates that change it.**
>
>     dS/dt = I(S) − E(S) at equilibrium: I(S*) = E(S*)
>
> In words: The change in species number per unit time is the immigration rate of new species minus the local extinction rate. Both rates depend on how many species are already there.
>
> - S: species now living on the island (richness) (species)
> - P: the species pool: species that could reach the island (species)
> - I(S): immigration: arrivals of species not already present, per unit time (species per day)
> - E(S): local extinction: residents lost from this island only, per unit time (species per day)
> - S*: the richness where the two rates are equal (species)
>
> Sanity check: On an empty island only immigration acts, so S rises. When all P pool species are present, immigration is zero and only extinction acts, so S falls.

In symbols, the film’s crossing S\* is where dS/dt = 0. It is **stable** (nudge richness away and it returns), and the turnover is I(S\*) = E(S\*).

The count wobbles around S\* because arrivals and losses are chance events. A smaller pool, which the film skipped, also lowers S\*, since immigration reaches zero sooner.

*Interactive on the lesson page:* Island equilibrium — immigration and extinction curves

## Working the numbers

**Worked problem.** The numbers are invented, not estimates for any real island. An island draws on a pool of P = 100 species. Immigration is I(S) = 10(1 - S/100) species per day and extinction is E(S) = 0.1S species per day: extinction is proportional to S, as when each resident has the same independent chance of loss. Assume continuous, deterministic change (no chance events), constant coefficients, no speciation, and an empty island at time zero.

1. *Equilibrium.* dS/dt = 10 - 0.1S - 0.1S = 10 - 0.2S. Setting this to zero gives **S* = 50 species*\*.
2. *Turnover.* At S = 50, I = 10 x 0.5 = 5 species per day and E = 0.1 x 50 = 5 species per day. Net change is zero, yet about **5 species are replaced each day** (10 events per day if you count gains and losses separately).
3. *Approach (calculus; optional).* The 0.2 per day in step 1 is a continuous rate constant (0.1 from slowing arrivals plus 0.1 from faster losses), so the gap to S\* decays exponentially, shrinking about 18% each full day (e−0.2 = 0.82). Solving from S = 0 gives S(t) = 50(1 - e−0.2t). Richness reaches 90% of S\*, 45 species, when e−0.2t = 0.1, so t = ln 10 / 0.2 = **11.5 days**. Check: at S = 45, I = 5.5 and E = 4.5 species per day, so richness is still rising by 1 species per day.

**Predict.** A smaller island has a lower S\*. Does it also turn over more slowly, or faster?

> **Check your prediction.** Faster, in the model. A smaller island is a steeper red line: double the extinction coefficient from 0.1 to 0.2. That lowers S\* to about 33, but about 6.7 species are then replaced each day, against 5: fewer residents, each lost more often. Nearness, by contrast, raises both S\* and turnover.
>
> | Island (one change from steps 1 and 2) | S\* | Turnover per day |
> | --- | --- | --- |
> | Original | 50 | 5.0 |
> | Nearer: I(0) = 15 | 60 | 6.0 |
> | Farther: I(0) = 5 | 33 | 3.3 |
> | Smaller: E = 0.2S | 33 | 6.7 |
> | Larger: E = 0.05S | 67 | 3.3 |

The table’s pattern is the model’s, not a measured one: real species differ in extinction chances and change one another’s chances, so real curves need not be straight.

The worked island, started empty, reaches about 47 species, 94% of S\*, within two weeks, though the count says nothing about which species. The rare real islands that are refilling, such as Krakatau, cannot be replicated or controlled, so the second film’s islands were emptied on purpose.

**Film transcript: Emptying Islands**

### Film transcript: Emptying Islands

The worked island refilled on paper, toward a steady count, while species kept coming and going. So empty a real island of every insect and spider. How many species come back, and are they the same ones? Most islands were colonised long ago, so nobody saw their rates from the start.

Daniel Simberloff and Edward Wilson chose islands small enough to empty, and then watched them fill. They used six small islands of red mangrove in the Florida Keys. Each carried a community of terrestrial arthropods: insects, spiders, mites and their relatives. First they recorded every arthropod species on each island. Then they covered each island with a tent and fumigated it. Afterwards they censused the islands again and again for a year. The count climbed back. By about day 250, most islands held about as many species as untreated islands. The exception was E1, the most distant.

Population densities, though, were still abnormally low. Now the second question. The count came back. How much of the old species list do you expect the new list to share? Take three mite kinds from island E9, chosen because their records are clear. One was there before and after. One was there only before. One was there only after. Two kinds before, and two kinds after, so the count is unchanged. But only one of the three is shared. The Jaccard similarity is shared species divided by species in either list. Here it is 1/3. Read it with care. These three are a chosen subset, not the whole community. And finding the shared mite at both censuses does not mean it survived.

It was recorded again after recolonisation. A second report followed four islands to the end of the second year. Species numbers had changed little from the year before, while immigrations and extinctions continued at high rates. A steady count with continuing replacement is what the model calls a dynamic equilibrium. The result fits the model, though on its own it does not prove the model's mechanism. And the rates themselves were rough. Censuses came at intervals, so some arrivals and losses fell between visits and were never seen. And a resident overlooked at one census looked like a loss and a return.

From the film

- Simberloff and Wilson censused the arthropods on six small red mangrove islands in the Florida Keys, removed them by fumigation under a tent, and censused again for a year.
- By about day 250, all but the most distant island, E1, held species numbers similar to untreated islands, though population densities were still abnormally low.
- The returning species were not all the ones present before, and in the second year arrivals and losses continued while the count changed little.
- The Jaccard similarity, species shared divided by species in either list, separates which species are present from how many.
- A steady count with a changed list fits a dynamic equilibrium without proving its mechanism, and census gaps make the measured rates rough.

## Testing it: emptying islands in the Florida Keys

*Figure (AI-generated illustration).* Emptying an island lets the model's rates be watched more directly than natural islands allow: census (count) the arthropod species (insects, spiders, mites and relatives), remove them by fumigation (gassing) under a tent, then census again as species return. Schematic: the tent is not drawn and the animals are generic. Image: three panels of one small red mangrove island. Panel 1, census before treatment: beetles, spiders and a red mite on the leaves beside an open notebook. Panel 2, fumigation: a pale cloud engulfs the island. Panel 3, censused again: a spider is back, and a moth, ant, grasshopper, wasp and beetle have arrived, beside a second notebook.

*How they came and went.* The authors inferred that most arrivals came through the air, by flight or carried on the wind. They also considered physical causes important in the extinctions of early colonists, so not every loss can be read as one species excluding another.

*Area by itself.* Large islands also differ in habitat, so in a 1976 study Simberloff censused eight mangrove islands, removed sections to reduce their area, and censused them again. Richness tracked island area independently of habitat diversity in that system, though the sample was too small to test whether smaller islands also turn over faster, as the table above predicts.

## Same number, different species

The second film’s three E9 mites shared one third of their list, and it was not just those three: in the 1969 report, the species on each island were not all the ones present before treatment. A caution the film did not give: dividing the mites’ changes by 364 days does not give a turnover rate, because other arrivals and losses between censuses are not counted. Is one third low for an island at equilibrium?

**Predict.** Make the worked island a chance process in which every species comes and goes independently of the others: each absent pool species arrives at a rate of 0.1 per day, and each resident is lost at a rate of 0.1 per day (continuous rates, as in step 3, not daily chances), every species alike. Census it at equilibrium, then 5, 10 and 100 days later. Does the shared fraction keep falling toward zero?

> **Check your prediction.** No. With equal rates in and out, each species is present half the time, and because species act independently, S wobbles around 50 (standard deviation about 5); species that came and went together would make the count swing far more. The expected Jaccard similarity falls from 1 to about 0.52 after 5 days and 0.40 after 10, then levels off near 0.33. It stops at a third because, long after, each species’ presence at the two censuses is like two independent coin flips: on both lists with chance 1/4, on neither with chance 1/4, so on at least one with chance 3/4, and 1/4 ÷ 3/4 = 1/3. With every species alike, a steady count sharing a third of its list is what this model expects; species present more or less than half the time give other values. The E9 three are hand-picked, so their one third is only a scale check.

For comparing communities by identity, see [diversity and succession](https://stemhop.com/lessons/diversity-and-succession/).

## Fragments, area and connectivity on land

A forest remnant surrounded by farmland looks like an island, and the species-area pattern is often summarised as S = cAz, where A is area and the fitted constants c and z set the overall level and how fast richness rises with area. The exponent z varies among taxa (groups of organisms) and sets of islands, so treat any single value as illustrative.

**Worked problem, continued.** Take z = 0.25.

1. *Habitat loss.* A habitat holding 50 species is cut from 100 to 10 km2. Richness is multiplied by (10/100)0.25 = 0.56, so it falls toward about 50 x 0.56 = 28 species. If the same curve holds for the remnant, this is the long-run value after **relaxation**, the slow loss of species that a smaller area can no longer hold, not a loss on the day of the cut. Treat it as a rough projection: some species survive in the surrounding land, and the method itself is debated (Going deeper).

Fragments are not true islands: the surrounding land, the **matrix**, is not open sea. Some species live in it, some cross it readily and some cannot cross it at all. That ties the model to the [**metapopulation**](https://stemhop.com/lessons/spatial-populations/) view: a set of local populations linked by dispersal, each subject to local extinction and recolonisation.

Rates can be a more sensitive signal than counts. In Swiss Jura grassland patches that Jasmin Joshi and colleagues isolated by mowing from 1993 to 1999, plant colonisation was 30% lower than in unisolated controls, yet after seven years the patches had only marginally fewer species, perhaps because counts lag behind rates, as in relaxation.

Connectivity has been tested over a longer span. In a longleaf pine savanna experiment reported by Ellen Damschen and colleagues (Science, 2019), patches joined by corridors had annual colonisation about 5% higher and annual extinction about 2% lower than unconnected patches across 239 plant species. After almost two decades, connected patches held nearly 14% more species and were still gaining; a five-year study would have seen much less of the effect. For reserve planning, see [conservation](https://stemhop.com/lessons/conservation/).

*Figure (AI-generated illustration).* A corridor's effect can be separated from the area it adds by giving unconnected patches dead-end side strips of equal combined area and matching edge-to-area ratio (how much border a patch has for its size), as in the longleaf pine savanna experiment (some controls were plain rectangles). Schematic, not the experiment's map: it gives each unconnected patch a single strip. Image: green habitat patches in a dark matrix. Left, two equal patches joined by a gold-outlined corridor. Right, two equal unconnected patches, each with a short dead-end side strip. An inset shows the corridor and the strips built from the same number of tiles.

### Design problem

A planner must choose between one 100 ha reserve and four 25 ha reserves, or decide whether to link existing patches with corridors. Design a study that tests whether connection itself changes the species found in patches.

> **One way to do it.** Island biogeography alone does not settle the reserve choice: one large reserve may hold more area-sensitive species, while several small ones may sample more habitat types and spread the risk of a local disaster. So test the mechanism.
>
> 1. *Replicate equal-area patches* in the same kind of habitat and matrix, and assign them at random to connected or unconnected treatments.
> 2. *Control for the corridor’s own area.* A corridor adds habitat and edge. Give some unconnected patches dead-end side strips matching the corridor’s area and edge-to-area ratio, and others the same extra area as a plain block, so a difference can be attributed to connection.
> 3. *Census species identity repeatedly*, at a fixed effort and season, for many years, recording which species are present as well as how many.
> 4. *Estimate colonisation and local extinction separately*, from changes in each species’ presence between censuses, and say how many events between visits you could have missed.
> 5. *Report both richness and composition* (for example Jaccard similarity to a reference patch). Effects can keep growing for years, although in the same system the effect on bird-dispersed plants had levelled off.
>
> The result applies to that habitat, matrix and set of species; whether it transfers to another matrix or to animals with other dispersal modes needs its own test.

## Going deeper

### Speciation and the book

MacArthur and Wilson also counted speciation on the island as a further source of species, which matters on large, remote islands, not small mangrove ones. Where it is appreciable, its rate adds to immigration, and the count settles where immigration plus speciation equals extinction. Their 1967 book, The Theory of Island Biogeography, developed the theory at length.

### The defaunation method

The islands were red mangrove, *Rhizophora mangle*, and each was tented and fumigated with methyl bromide; the full one-year report appeared in Ecology in 1969, and a second report in 1970 followed four islands to the end of the second year (the steady count the second film ends on). The 1969 appendix marks each record as seen, inferred present or inferred absent; the film’s E9 mites were chosen for unambiguous records. Treatment had its own side effects: on island E7, fumigation killed 85% of the mangrove tree.

### How rough the measured rates were

Most observed immigration and extinction rates lay between 0.05 and 0.50 species per day, against about 0.1 to 1.0 from a simple noninteractive model (species do not change one another’s chances). The authors suggested many events between censuses went unrecorded.

### Running a species-area curve backwards

Step 4 used a species-area curve in reverse, predicting losses from a shrinking area. Whether that overestimates extinctions after habitat loss is debated.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/island-biogeography/
