Biogeography is the study of the temporal and spatial distribution of plants and animals across the planet. Unlike traditional community ecology, biogeography deals with a much larger scale in both time and space, considering issues such as migration, extinction, dispersal, and speciation (Hubbell 1997). Island biogeography is concerned with how these processes occur in insular areas, as exemplified by true oceanic islands. This paper seeks to present the principle concepts of island biogeography and to discuss their application to the design of nature reserves, which increasingly function as isolated islands of habitat within seas of human-dominated landscapes.
PRINCIPLES OF ISLAND BIOGEOGRAPHY
One basic assumption of island biogeography is that no island will contain the species diversity of the mainland which served as the source for colonisation (MacArthur 1972). This has been supported by numerous studies, most commonly among bird species in California (MacArthur 1972). The reason is fairly clear: To disperse to an island, a species must successfully cross an expanse of water. Water is obviously most limiting to terrestrial species, least limiting to aquatic species, and of very little importance one way or the other to wind-dispersed species (Hunter 1990). Regardless of species type, getting to an island involves traversing distance as well as water.
In a critical paper published in 1967, R.H. MacArthur and E.O. Wilson presented the theory of equilibrium for island biogeography. Stated simply, the pair postulated that an island reaches dynamic equilibrium in population size and species diversity through the processes of migration and extinction. New species or members arrive from the source area (the mainland) and existing species become extinct from competition, natural disaster, or genetic inbreeding. In this theory, the adaptation and evolution of species is assumed to progress at rates so slow in comparison to rates of extinction and immigration that it can be safely ignored (Johnson et al. 2000). Evolution is a significant process only on islands that are large and stable; otherwise, species simply don’t survive long enough to undergo natural adaptation (MacArthur & Wilson 1967). At any one time after equilibrium is achieved, the island will sustain the same number of individuals and species; however, the composition of species will change. The two most important criteria for determining the rate of species change are the area of the island and its distance from the mainland source. Two related but less important factors are connectivity to the mainland and circuitry with other islands.
1. Island area
The concept that small area equals low species diversity is well-supported, although the connection is not actually that direct (MacArthur & Wilson 1967). The true progression may be stated thus: smaller area = less habitat diversity = less species diversity. In addition to lowering habitat diversity and therefore species diversity, small areas support fewer members of any one species. Small populations are more prone to extinction, since they are proportionally more affected by problems such as natural disaster, sex/age fluctuations, environmental change, and genetic inbreeding (Hunter 1990). Therefore, biogeographers generally agree that smaller islands are less capable of supporting high levels of population density and species diversity.
2. Island isolation
The relative isolation of an island from its mainland species source affects the rate of successful immigration. A large island close to the source will sustain more diversity than a identically-sized island far from the source simply because more species reach it (MacArthur & Wilson 1967). Conversely, a large island far from the source may have an immigration rate so low that evolution begins to shape the biota, creating entirely new species (Hubbell 1997). These conclusions hardly require a Ph.D. to ascertain, but the issue is somewhat more complex. Isolation is also affected by how the island is oriented in relation to prevailing winds, currents, and storms (Dramstad et al. 1996). Some Pacific atolls seem tantalizingly close to total isolation because they lay outside the path of immigrant-bearing currents (Davidson 1998). A number of natural occurrences serve to mitigate the effects of island isolation from the mainland; together, these are termed connectivity.
The degree to which an island is connected to the mainland can mediate the factor of isolation. The two most common connectors for islands are land bridges and stepping stones.
For a terrestrial species, travel across an expanse of ocean is a risky endeavour. However, paleogeography has demonstrated that during key moments in earth history (such as ice ages), sea level has dropped sufficiently to form land bridges connecting formerly or currently separate land masses. For most Americans, this is typified by the Siberia-Alaska land bridge which shuttled plants, animals, and humans into the Western Hemisphere. For tropical ecologists, the more interesting effects of this same glaciation were the connection of Indonesia with mainland Asia and Papua New Guinea with Australia, providing corridors for species from one area to the other (Davidson 1998). Land bridges form the most easily traversable terrain for immigration of terrestrial species.
In considering the dispersal of species from the mainland to islands, even a layperson can readily distinguish that “stepping stone” islands exist, providing not quite a land bridge but certainly more of a fighting chance than a solid expanse of ocean. Stepping stones are intermediary between a full land bridge and a solid stretch of ocean (Dramstad et al. 1996). If an island is separated from the mainland by a series of other islands, this will increase the immigration rates of certain species, most notably birds and mammals (MacArthur & Wilson 1967). Passively-dispersed species, such as micro-organisms and many plants, rely more on wind power and therefore do not show a significant change in immigration rates with stepping stones (MacArthur & Wilson 1967). The effectiveness of a stepping stone is determined by three factors: the size of the stepping stone, the size of the recipient island, and the distance between them (MacArthur & Wilson 1967). Of these factors, the most important is the distance separating them (MacArthur & Wilson 1967). For highly visual species, such as many predators, the next stepping stone must be within sight to entice a member to make the journey (Dramstad et al. 1996). But for most species, the journey is made by chance or desperation, and distance functions primarily as a determinant of how many members survive to the next patch of land. As species traverse the series of stepping stones, some species will not survive to the next island. In this way, isolation from the mainland remains a key determinant of immigration despite the advantage of stepping stones as intermediaries. Another consideration is that as the number of alternate stepping stones increases for each recipient island, the flow of species from the source increases (MacArthur & Wilson 1967, Dramstad et al. 1996). Or, visually:
MAINLAND -----> STEPPING -----> RECIPIENT
MAINLAND -----> STEPPING STONE -----> RECIPIENT
-----> STEPPING STONE -----> ISLAND
-----> STEPPING STONE ----->
Another important mediating factor on the effects of island size and isolation is the circuitry that exists within an island-mainland system. Circuitry refers to the degree to which loops or alternate routes are present within a system (Dramstad et al. 1996). Alternate routes increase the chance that a species will successfully immigrate to an island. For example, if a natural disaster kills an entire bird generation on one island but a nearby island also supports a population of the species, some members will survive to emigrate to the next island in the system. It is possible for a network to be connected but not circuitous, as in this example (from Dramstad et al. 1996):
X --- X --- X X ------------ X
| | |
X X ------------ X
high connectivity high connectivity
low circuitry high circuitry
APPLICATION TO NATURE RESERVES
From the beginning, island biogeographers identified non-oceanic islands, such as the isolated tops of mountains in North America, as functional habitat islands. However, only recently have researchers accepted the idea that nature reserves may become so isolated from similar habitats that they function as islands (Dramstad et al. 1996). At what point does human-dominated land become as effective a barrier to dispersal as an ocean? In 1967, pioneers MacArthur & Wilson did not consider this possible:
...these [non-oceanic] islands are different in that the space separating them is not barren of competitors. A true oceanic island clad in spruce forest would have few bird species present, but those would be typical of spruce forests; a small patch of spruce amid southern deciduous forests might be well populated mostly with deciduous forest species which had overflowed from adjacent habitats.
(MacArthur & Wilson 1967, p. 114)
In a later publication, MacArthur reiterates that regardless of habitat degradation, dispersal over land is less hazardous than dispersal over water, at least for primarily terrestrial species (MacArthur 1972). While researchers may disagree as to the extent that reserves follow the patterns of islands, island biogeography is becomingly increasingly important in the design of nature reserves throughout the world.
1. Reserve size
A common debate, often considered settled and just as often resurrecting itself over some new bit of research, is the question of whether one large, undisturbed tract of land is preferable for a nature reserve, or, alternately, several small patches that add up to an equal area. The debate has gained a popular acronym -- SLOSS, which stands for Single Large Or Several Small. If we consider reserves as islands, then the species-area relationship, as well as the immigration-extinction curve, argues forcefully for a single large reserve (Noss et al. 1997). However, the issue is far from settled. A study of species distribution for both habitat and oceanic islands found that several small connected islands usually had higher species diversity for plants, invertebrates, reptiles, and amphibians; while a single large island usually had higher species diversity for birds and mammals (Boecklen 1997). A study of squirrel, mice, and chipmunk species in Indiana found that increasing forest size was directly proportional to species diversity (Nupp & Swihart 2000). Conservationists have responded by advocating a variety of reserve sizes. The Nature Conservancy warns that it is better to reserve the large area now and later break it up than to wait for more research and thereby lose all hope of gaining large tracts of land in the future (Noss et al. 1997).
2. Reserve isolation
The relative isolation of a reserve compared to a true oceanic island is at the heart of current debate. Of critical importance is the matrix, the predominant material in which a patch (be it reserve or island) is embedded, and the contrast between the matrix and the patch (Dramstad et al. 1996). There is a sharp contrast, for example, between a patch of volcanic island and the surrounding matrix of deep ocean. Less sharp is the contrast between an old-growth forest and a meticulously groomed park. But contrast varies by species. At least two studies have found that many small mammals will not traverse even a narrow, grassy path that is devoid of traffic (Hunter 1990). And yet a mountain lion was found lurking in the downtown park of Victoria, British Columbia, a city of some 330,00 inhabitants. As Noss et al. wrily point out, “Many organisms are capable of crossing narrow swaths of unsuitable habitat, such as a trail, a narrow road, or a vacant lot; far fewer are able to successfully traverse a six-lane highway or the City of Los Angeles” (Noss et al. 1997, p. 99). In fact, studies indicate that for most species reserve isolation is detrimental only in areas where the matrix is fundamentally different from the patch (Noss et al. 1997). On the other hand, the species that are significantly affected by even a slight contrast between matrix and patch tend to be the high-quality interior species, those that cannot abide edge effects (Hunter 1990).
A common technique to minimize contrast between matrix and patch, and therefore to increase the success of migrations, is to provide a buffer zone around the reserve . The buffer zone is carefully managed to allow low levels of human use, such as small-scale agro-forestry, eco-tourism, or selective logging (Noss et al. 1997). This zone softens the transition from human-dominated landscapes to the nature reserve and provides a more amenable matrix for species to cross. When designing reserves, conservationists should identify the dispersal tactics of the targeted species and use this to predict what type of matrix and what distance of this matrix each species will be able to cross.
In island biogeography theory, islands are isolated from a mainland source, which constantly sends immigrants to the island. In a world where nature is increasingly confined to islands, we must at some point ask the question, “What is the mainland source?” Many scientists doubt that even a large reserve can exist in isolation from a vastly larger source of genetic material (Hunter 1990). In answer to this, planners have stressed the need for providing connectivity and circuitry among nature reserves.
In areas where matrix and patch are fundamentally different, or at least vary enough to inhibit species movement, reserves should be intentionally connected. Connections are critical to wildlife health because they allow for genetic sharing among populations, which in turn increases species’ likelihood of survival. In addition, connectivity can effectively unite many small patches into one large habitat, suitable for wide-ranging species such as wolves and panthers (Noss et al. 1997). As with true oceanic islands, connections that are commonly effected for nature reserves can be corridors or stepping stones.
Corridors are the terrestrial equivalent of land bridges for oceanic islands. A corridor is an unbroken pathway from one reserve to another. They may be as narrow as a fence line, for small mammals such as chipmunks; or many kilometers wide, for interior species such as trees. If a species cannot make the complete journey in its lifetime, the corridor must provide suitable living habitat so that the plant or animal can breed and move in generations along the corridor to the next reserve (Hunter 1990). This particular point is especially crucial for perennial plants and small vertebrates. Corridors succeed when they are tailored to fit the needs of the target species; unfortunately, too often economics dictates a narrow corridor that is entirely edge, preventing the movement of a great many species (Dramstad et al. 1996).
Again borrowing from island biogeography, an alternate possibility for connecting reserves is to use stepping stone patches between large reserves. A stepping stone is essentially a “rest stop” within the matrix which improves the chances of successful immigration. Although less effective than a fully-functioning corridor, stepping stones may be preferable where cost or ownership issues dictate a very narrow corridor with no interior core. For most species, stepping stones are chanced upon in the process of immigration, but as noted earlier, highly visual species will make use of them if the next stepping stone is within sight (Dramstad et al. 1996). Again, it is important to consider the dispersal mechanisms of each species. For example, a plant whose seeds can travel up to 50 km has different needs than a plant whose seeds can travel only 5 km; different still is the plant that relies on a certain animal to move its seeds. All of these issues must be considered when planning stepping stones. If interior species are to move, each stepping stone must provide some interior core habitat.
A group of nature reserves may be very well connected through the use of corridors or stepping stones, but planners must also provide for circuitry. If the reserves are linked in more or less a straight line pattern, the movement occurs in only two possible directions. However, if the reserves are clustered and linked with many loops or alternate routes, movement can occur in many different directions (Dramstad et al. 1996). Increased circuitry speeds the interaction between patches, which in turn allows for more effective genetic sharing and a more realistic simulation of natural ecosystems (Noss et al. 1997). However, by decreasing barriers to movement, circuitry also provides pathways for catastrophes such as fire, pests, and disease (Hunter 1990). Ideally a reserve or a connected/circuited series of reserves should be large enough to allow for these natural disaster regimes, as most natural ecosystems do (Hunter 1990). For example, if natural forest fires generally burn about 1,000 ha, the reserve should be larger than 1,000 ha. Anything less is inviting catastrophe. In situations where it is impossible to provide enough space to allow for natural processes, planners must decide how much circuitry to allow and how much human interference to provide in the case of natural disaster. This should be only a short-term answer to sustaining life within the reserve; the ultimate goal should be to rehabilitate degraded habitat in order to increase the reserve to a functional area (Noss et al. date?).
Concepts and research of island biogeographic theory have had a significant impact on the planning and management of nature reserves. Strong links exist between islands and reserves in terms of area, isolation, connectivity, and circuitry. Except in special cases, conservation biologists now recommend that reserves be composed of a single large area. When this is not feasible, several small patches can be substituted, but care must be taken to provide high degrees of connectivity and circuitry. Connections, either by corridors or stepping stones, must be planned based on the dispersal mechanisms of target species. In all cases, the matrix surrounding the reserve should be managed to minimize contrast, such as the use of buffer zones. And finally, one primary divergence from island biogeography remains to be answered: If we are confining nature to islands, what is the “mainland” source for immigration?
Boecklen, W.J. Nestedness, biogeographic theory, and the design of nature reserves. Oecologia (1997) 112: 123-142.
Davidson, O.G. The Enchanted Braid: Coming to Terms with Nature on the Coral Reef. Wiley and Sons (New York, NY). 1998.
Dramstad, W.E.; Olson, J.D.; and Forman, R.T.T. Landscape Ecology Principles in Landscape Architecture and Land-Use Planning. Island Press (Washington, DC). 1996.
Hubbell, S.P. A unified theory of biogeography and relative species abundance and its application to tropical rain forests and coral reefs. Coral Reefs (1997) 16, Suppl.: S9- S21.
Hunter, M.L. Wildlife, Forests, and Forestry: Principles of Managing Forests for Biological Diversity. Prentice-Hall (Englewood Cliffs, NJ). 1990.
Johnson, K.P.; Adler, F.R.; and Cherry, J.L. Genetic and phylogenetic consequences of island biogeography. Evolution (2000) 54, vol.2, 387-396.
MacArthur, R.H. Geographical Ecology: Patterns in the Distribution of Species. Harper and Row (New York, NY). 1972.
MacArthur, R.H. and Wilson, E.O. The Theory of Island Biogeography. Princeton University Press (Princeton, NJ). 1967.
Noss, R.F.; LaRoe, E.T.; and Scott, J.M. Endangered ecosystems of the United States: a preliminary assessment of loss and degradation. U.S. Geological Survey publications, Biological Resources Division. No date given for publication, but latest reference cited was 1998.
Noss, R.F.; O’Connell, M.A.; and Murphy, D.D. The Science of Conservation Planning. Island Press (Washington, DC). 1997.
Nupp, T.E. and Swihart, R.K. Landscape-level correlates of small-mammal assemblages in forest fragments of farmland. Journal of Mammology (2000) 81, vol. 2: 512-526.
Return to Topic Menu
It is 6:22:03 AM on Friday, June 23, 2017. Last Update: Saturday, May 4, 2002