Research on islands has long played a fundamental part in developing our basic understanding of ecology and evolution. Both Darwin's and Wallace's insights into evolution and speciation were shaped by studies on islands (Darwin 1859; Wallace 1881). It is no coincidence that Darwin felt close to "that great fact - that mystery of mysteries - the iirst appearance of new beings on this earth" in the Galapagos (Darwin 1845). Even today, the Hawaiian drosophilids provide a primary standard for analyses of speciation (Carson et a!. 1970; Kaneshiro 1995). More recently, ecology has been enriched by analyses of competition and character displacement (Lack 1947; Brown and Wilson 1956) and by island biogeography theory (MacArthur and Wilson 1967), which were developed and tested on islands, but have been applied much more widely.
The reasons why islands are useful in ecological studies are straightforward. island populations, communities and ecosystems arc self-maintaining entities with well-defined geographical limits that contain the fundamental processes, properties and interactions of ccological systems - but they often do so in a simpler way, without the complexity of most continental systems. Moreover, the influence of particular factors that control ecological phenomena can often be understood against a relatively simple background in island systems. For example, the evolutionary radiation of a group of plants or animals that resulted from a single founding population can be traced with a certainty rarely achievable on continents or continental islands (Eliasson 1995, Kaneshiro 1995); the influences of immigration/establishment
Functional Rotes of Biodiversity: A Global Perspective
Edited by H.A. Mooncy, J.H. Cushman. E. Medina, O.E. Sala and E.-D. Schuize (("3$) © 1996 SCOPE Published in 1996 by John Wiley & Sons Ltd **
versus extinction on species richness can be analyzed - if not with certainty, then at least with a clearer focus than is possible on continents (Adsersen 1995, Roughgarden 1995), and even the individual factors controlling ecosystem biogeochemistry can be analyzed in isolation to a degree scarcely dreamt of in continental systems (Vitousek el al. 1992, 1994; Vitousek and Benning 1995).
This is not to say that the results of research on islands can be applied directly to continental systems; the greater taxonomic complexity of most continental systems, and perhaps qualitative differences between continental and oceanic island systems (Roughgarden 1995), may defeat any simple extrapolation. Rather, the understanding of ecological and evolutionary processes gained on islands can be used to identify important processes, support the development of theory, and test the limits of conceptual and mathematical models. The understanding gained by the use of those processes, theories and models can then be applied in a more complex continental context.
More practically, islands provide a record of humanity's interactions with biological diversity in contained areas, and of the consequences of those interactions. Because the modern epidemic of anthropogenic extinctions has hit first, and hardest, on oceanic islands, islands also provide a set of management experiments in which the consequences of different approaches to managing populations, species and ecosystems that are now on the verge of extinction can be evaluated (MacDonald and Cooper 1995). Lessons learned from succcsscs and failures in conserving the biological diversity of islands might guide us in developing strategies for protecting continental diversity.
Can the same features of island ecosystems that have proved useful to our understanding of ecology and evolution, and that may help us to manage threatened populations and ecosystems, also contribute to understanding the interactions between biological diversity and ecosystem function? We believe so; in fact, wc believe that occanic island ecosystems are particularly useful for studies of species-level diversity and ecosystem function. First, island habitats generally contain fewer species than comparable continental habitats, so that experiments that manipulate diversity may be more manageable. Second, invasions and extinctions are widespread on islands, and they can be used to evaluate the effects of inserting and deleting species and/or functional groups. Third, and most interestingly, diversity varies among islands and between islands and continents for reasons that differ in part from the determinants of variation in species diversity on continents.
The major reasons for variation in continental systems can be found in climate (especially the great latitudinal pattern from the arctic to the tropics), in disturbance, in soil fertility and in other factors. However, the same factors that affect diversity also affect numerous ecosystem functions strongly and directly, and teasing apart the resulting interactions is not easy, indeed not often possible (Vitousek and Hooper 1993). Even in the cases where variation in diversity on continents is not tied dircctiy to known environmental factors, it is difficult to be confident that both diversity and any related pattern in ecosystem function are not both controlled by another factor. (Those concerns do not apply to experiments in which diversity is deliberately manipulated as a factor, only to the many cases in which we attempt to use biogeographic patterns, dynamics or experiments that were designed for other purposes to identify connections between diversity and ecosystem function.)
In contrast, island ecosystems vary in diversity not only for the reasons described above, but also, strikingly, because of the rarity of successful colonization and establishment on remote oceanic islands (and also because of island size and the time over which colonization and speciation have operated). It is possible to loeate islands in which population or species-levei diversity is very low. under tropical climatic conditions where diversity would be substantial on a comparable area of a continent or continental island. It is also possible to locate situations in which climate and soils generally are similar on a range of islands that nevertheless differ in diversity owing to their distance from a source area. For example, in the southwest Pacific the richness of mangrove taxa varies with distance from their regional center of diversity, ranging from 30 in Papua New Guinea to no native species in the Society Islands (Figure 10.1) (Woodroffe 1987). Finally, ecosystem properties and processes themselves often vary in more comprehensible ways on many islands that in most continental situations (Vitousek and Benning 1995), and many ecosystem studies on continents owe their success to their choice of systems with well-defined geographical limits (cf. Likens et al. 1977). Consequently, both sides of the biological diversity/ecosystem function interaction may be relatively approachable on islands.
Research on islands can also be useful for evaluating the utility (even the reality) of defining functional groups of species in analyses of biological diversity and ecosystem function (Körner 1993; Meyer 1993; Solbrig 1993). Functional groups are defined here as sets of populations or species that affect or control an ecosystem-level process in similar ways; they are analogous to guilds in community ecology, and may indeed overlap with them, depending upon the function under study. Islands are useful for studying functional groups bccause the remoteness of oceanic islands acts as a selective filter; islands often lack whole groups of species that play important functional roles in continental ecosystems (e.g. ants, termites, grazing mammals). Accordingly, the biota of oceanic islands is considered to be "disharmonic" (Carlquist 1965, Eiiasson 1995). Often other groups of organisms develop in unusual ways and fill the ecological role of an absent
group. One spectacular example is the evolution of large, flightless herbivorous birds on many oceanic islands (James 1995). Often these substitutes are not able to maintain themselves in the face of changes set in motion by the arrival of humans on an island - and while the loss of these spectacular examples of evolutionary radiation on islands is tragic, the accompanying rearrangements in ecosystem function could at least yield insight into interactions between diversity and ecosystem function.
This last example also illustrates some of the difficulties of studies on island ecosystems. First, they are not just less diverse than continental ecosystems, their disharmony may also make them different in some fundamental ways (Roughgarden 1995). Second, few islands are unaltered by humanity, and often anthropogenic alterations have taken place on a scale even greater than that in most continental systems. The disproportionate representation of island species in the lists of threatened and endangered species and of recent extinctions makes one consequence of those changes clear (MacDonald and Cooper 1995). Consequently, to evaluate biological diversity and eco,system function on islands, we have to deal with remnants (themselves altered) of the systems that were once there, and/or learn to read the historical record with a precision that has hitherto been difficult to achieve (Maunder et al. 1995). Nevertheless, even in their altered state, islands offer a range of systems in which diversity varies independently of other controls on ecosystem function; they therefore offer great opportunities for observation, analysis and experiment.
Was this article helpful?