Biodiversity island models

7/21/2019 Biodiversity island models

http://slidepdf.com/reader/full/biodiversity-island-models 1/13

OCEANIC ISLANDS:

MODELS OF DIVERSITY

Rosemary G. GillespieUniversity of California, Berkeley

I. IntroductionII. Characteristics of Biodiversity

III. Island DynamicsIV. Species LossV. The Future of Biodiversity on Oceanic Islands

GLOSSARY

adaptive radiation Evolution of ecological and phe-

notypic diversity within a rapidly multiplyinglineage.

allopatric Occurring in separate, nonoverlappinggeographic areas.

dioecious In plants, having the male and femalereproductive organs borne on separate individualsof the same species.

disharmony The nonrandom representation of biotaamong natural colonists of oceanic islands as com-pared with mainland systems. Species that are moredispersive will have a higher representation in thebiota, the effect being accentuated by the tendency

of these few successful colonists to diversify intomultiple species through adaptive radiation.

ecomorph A local population or group associatedwith a particular ecological niche; the terms has nosystematic or taxonomic significance.

endemic/endemism Unique to a given locale; occursin the site at which it is endemic, and nowhere else.

impoverishment On an island, paucity of speciesrelative to adjacent mainland.

invasional meltdown The compounded effect on anecological community of successful and successive

invasions of alien species benefiting from theprevious introduction of the other.

microcosm A small, representative system that hasanalogies to a larger system. Oceanic islands areoften considered microcosms of continental areasbecause the composition of species on islands, andthe processes of formation and decline, are analo-gous but occur on a very small scale.

parapatric Occurring in geographical areas that abutbut do not overlap.

phylogenetics The study of evolutionary relatednessamong various groups of organisms (e.g., speciesand populations). Phylogenetics, also known asphylogenetic systematics, treats a species as a groupof lineage-connected individuals over time.

relictualization Survival of species on islands afterthose on adjacent mainland areas have sufferedextinction.

sympatric Occurring in the same geographic areaswithout interbreeding.

taxon cycle A hypothesis proposed by E. O. Wilsonto describe the common niche expansion amongcolonizers, and subsequent long term toward dif-

ferentiation, with isolated species narrowing theirniches and restricting their ranges resulting ineventual extinction.

OCEANIC ISLANDS ARE UNUSUAL IN THEIR ISO-

LATION , which has allowed the formation of uniquelyevolved biotas that exhibit community characteristicssuch as disharmony, high endemism, and sometimes

1

Encyclopedia of Biodiversity

Copyright & 2007 Elsevier Inc. All rights of reproduction in any form reserved.



relictualism. Characteristics of species may includereduced dispersal capabilities, evolutionary innovat-ions, and size and reproductive changes. Ecologicaland evolutionary changes play a key role in successfulisland colonization, and often result in adaptive radi-

ation. While such circumstances may have facilitatedrapid speciation, they also result in high rates of extinction, greatly accelerated since the arrival of humans. Human-mediated impacts on islands cannotbe held back; however, the islands present a relativelysimple system in which it is possible to understand thedynamics of species formation and loss.

I. INTRODUCTION

There is a huge diversity of types of islands, but the

common denominator is that they are isolated, well-defined geographically, and have distinct boundaries.These characteristics can result in properties such as of a microcosmal nature and a uniquely evolved biota.Indeed, the flora and fauna of oceanic islands inspiredscientific interest from the moment explorers firstbegan sailing the oceans. The importance of evolutio-nary processes on islands was first recognized inCharles Darwin’s work on the Galapagos, which gaverise to the theory of evolution by means of naturalselection. Likewise, Alfred Russell Wallace contempo-raneously developed similar theories based on studies

in the Spice Islands of Indonesia. Over the last century,research has focused on multiple facets of islands—from the mechanism of formation of the isolated hab-itat and how its geological history has been shaped, tothe colonization of species and the development of uniquely evolved biotas in similarly unique commu-nities and ecosystems (Keast and Miller, 1996). Islandecosystems tend to be simple when compared withcontinental ecosystems; thus, the way ecosystemsfunction can be more easily studied. In essence, is-lands can be considered as Nature’s test tubes. Eachisland represents a trial in an experiment and each new

island is the repeat of one of these experiments.Islands can be broadly classified as one of the two

types, continental or oceanic. Continental islands arefragments of a larger landmass, often created by a risein sea level or when land breaks off and drifts from themainland. Oceanic islands (e.g., Fig. 1) can result fromvolcanoes rising above the water on or near a mid-ocean ridge, or where lithospheric plates converge(Nunn, 1994). The key biological difference between acontinental and an oceanic island is that the former,when created, has a full complement of species (or at

least a sample of what occurred on the continent be-fore the fragment became isolated), and the number of species, at least initially, will tend to decline as a resultof stochastic events. In contrast, oceanic islands areformed without life, so the number of species can only

increase, and will do so at a rate that will depend onthe interplay of isolation and time (Gillespie andRoderick, 2002). If the islands are not extremely iso-lated, equilibrium will be reached between immigra-tion of colonists and extinction. With greater isolation(and lower immigration rates), and given sufficienttopographical diversity, local endemism will increaseover time. Extreme isolation, with the frequency of colonization being insufficient to allow the occupationof available ecological niche space through migration,can provide the opportunity for adaptive radiation,and has been called as the ‘‘radiation zone’’ (Whittaker,

1998). Here, species diversity arises through the for-mation of multiple new species adapted to exploit theavailable ecological space.

Over the last few years, geological understanding of islands has improved immensely. For many oceanicarchipelagoes, the age of each island is often knownwith some level of precision. This, coupled with thefact that diversification on these islands has a clearlyrecognizable beginning (colonization from outside),has given biologists a temporal framework withinwhich to examine ecological and evolutionary proc-esses, and the formation of communities over time.

II. CHARACTERISTICS OF BIODIVERSITY

A. Community Characteristics

1. Disharmony

On oceanic islands, the biota is often considered to be‘‘disharmonic.’’ This phenomenon arises partly becausetaxa differ considerably in their ability to colonize re-mote islands, leading to attenuation in the number of groups represented on more isolated islands. More-

over, as a consequence of the isolation, islands gene-rally show a degree of impoverishment whencompared with continental communities. Adaptive ra-diation accentuates this effect, with multiple speciesarising from the few that successfully colonize, theresult being very few families and genera comparedwith overall diversity. This pattern has been noted inthe Canary Islands, (Juan et al., 2000) but the effect isparticularly well illustrated as one crosses the PacificOcean from west to east in the direction of increasingisolation of an oceanic island (Fig. 2; it should be

2 _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________



noted that this figure excludes continental islands suchas New Caledonia and New Zealand, in which ancientconnections to continental landmasses also appear tohave allowed high levels of endemism).

2. Endemism

The pattern of species accumulation on oceanic is-lands depends on the rate of evolution relative to thefrequency of island colonization, the extreme cases

leading to extensive adaptive radiation as a result of insitu evolution with associated adaptation to occupy theavailable ecological space. On remote islands, the fre-quency of colonization becomes vanishingly rare. As aresult, the biodiversity of remote islands has largelyarisen through evolution and adaptation of the fewinitial colonists, and levels of endemism are conse-quently very high (see Table I for a comparison of endemism among different groups of species in the

FIGURE 1 Comparison of oceanic islands. (a) View from the top of Mt. Rotui, Moorea (Society Island archipelago, French Polynesia)

showing topographical diversity characteristic of younger volcanic islands; higher elevation forested habitat is shown in the foreground.

(b) Atoll in Micronesia. Such oceanic islands, which show little topographic diversity, harbor relatively few terrestrial species, with little

endemism. Photos by G. K. Roderick and R. G. Gillespie.

_____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________3_____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________3



Canary, Galapagos, and Hawaiian islands). The effectof isolation on endemicity is shown in Fig. 2, whichillustrates how endemism increases with isolation in

the oceanic islands of the Pacific.

3. Relictualism

The mainland taxa that give rise to the island colonistsexist in an environment that is often very differentfrom that of species on islands. Accordingly, speciesthat colonize islands may suffer extinction, though thiscannot generally be distinguished from the effects of disharmony. Likewise, island progenitors may becomeextinct on the mainland yet remain extant on islands,

because competition may not be as severe, the climateis less extreme; in this case the species on the islandscan be considered ‘‘relicts.’’ Clearly, this effect will bemuch more pronounced on older islands or archi-pelagoes.

B. Species Characteristics

1. Reduced Dispersal Capabilities

Loss of dispersal ability is a common feature of taxa onoceanic islands (Williamson, 1981). The phenomenonwas first documented by Darwin, who suggested that

‘‘powers of flight would be injurious to insects inhab-iting a confined locality, and expose them to be blownto the sea.’’ For example, island plants are renownedfor their large seed size and associated loss of dispersalability. Among endemic/indigenous insects, 90% of those on Tristan de Cunha and 40% on CampbellIsland show some degree of wing reduction. Amongbirds, flightlessness has evolved repeatedly on islandswhere predators are absent. Such reduction in dispersalability may accelerate species diversification. However,many of the flightless birds on islands have been driven

0

500

1000

1500

2000

2500

3000

3500

Solomons Vanuata Wal lis &

Fatuna

Tonga Kiribati Ind. Samoa Am. Samoa Cook Is. Societies Marquesas Hawaii

0

10

20

30

40

50

60

70

80

90

100

Indigenous spp.

Area

% endemism

A r e a / n o .

i n d i g e n o u s s p e c i e s

% en d emi s m

Hawaii

Marquesas

French PolynesiaSocieties

Cook Is

Am.Samoa

W.Sam.

Tonga

Wallis Is.

Fiji

Solomon ls.

Vanuatu

Papua

New Guinea

Australia

FIGURE 2 Plant transect across selected Pacific islands: indigenous species, area, and endemism. Larger islands have dis-

proportionately larger numbers of species, as documented in MacArthur and Wilson (1967)’s ‘‘Equilibrium Theory of Island

Biogeography.’’ The initial idea of equilibrium numbers of species on islands was based on a balance between immigration

(decreases with distance from a source of colonists) and extinction (decreases with island size). However, the high levels of

endemism on remote islands also reflects the role of evolution within these islands, which may compensate for low levels of

immigration with numerous species arising through adaptive radiation.

TABLE I

Approximate levels of endemism (%) for different groups of native

organisms on three of the better-known oceanic archipelagoes

Birds (all

species)

Flowering

plants

Terrestrial

invertebrates

Canary Islands 45 40 58

Galapagos Islands 48 36 50

Hawaiian Islands 81 91 98




to extinction as human occupation spread, for exam-ple, in Hawaii the flightless rails and geese and thelarge flightless waterfowl the moa nalos, in Mauritiusthe Dodo, and in Madagascar the elephant bird.

2. Innovations

The known disharmony of islands appears to stimulatethe development of some unusual traits in islandgroups. For example, Hawaiian birds have developed aremarkable diversity of elongate and decurved billshapes, presumably because of the availability of nec-tar resources. The Hawaiian silversword plant allianceincludes trees, shrubs, subshrubs, rosette plants, cush-ion plants, and a vine, that occur in some of the wet-test sites recorded on Earth to extreme desert-likeenvironments. Among insects, the absence of native

social insects in some of the more remote islands of thePacific may have been largely responsible for the ad-aptive shift to predation in some terrestrial insectgroups and the diversity of other predatory arthropodson these islands. One of the most striking innovationsis the development of predatory behavior in a lineageof Hawaiian caterpillar moths in the genus Eupithecia(Howarth and Mull, 1992). Another innovation inHawaii has been the development of odonate damsel-flies with terrestrial larvae, associated with the paucityof lakes in the upland yet moist forests of the islands.Similarly, the unusual habitat of cryptogram herbivory

in beetles in the subantarctic islands is considered tobe a result of the disharmony of the flora.

3. Size Changes

Species on islands show a tendency toward size ex-tremes, both gigantism and dwarfism. Overall, amongmammal species that colonize islands, large speciestend to get smaller while small ones tend to get larger.Included among the islands’ giants are the giantflightless birds of New Zealand (moa) and Hawaii,giant sheet-web spiders of the remote Pacific islands,and hissing cockroaches of Madagascar. Dwarf species

include elephants, foxes, rabbits, and snakes as well asthe pygmy mammoth, which once inhabited Califor-nia’s Channel Islands. It appears that size changes areassociated with the very different conditions experi-enced by island species, often with fewer constraintson size extremes.

4. Reproductive Changes

A common pattern seen in plants that colonize islandsis the development of dioecy, asexual reproduction, and

low rates of reproduction, each of which may arise as aresult of selection for loss of dispersal ability, togetherwith high selfing rates, after lineages have become es-tablished on remote islands. The Hawaiian Islands, forexample, are particularly rich in dioecious species with

about 15% dioecy in native species of flowering plants(compared with 6% in mainland species).

III. ISLAND DYNAMICS

A. Colonization

Given a suitable abiotic environment, the success of any given colonization event depends largely on theavailability of ecological space, although chance clearlyplays a role. The importance of available ecological

space is clearly illustrated on isolated archipelagoesthat are formed serially, some in a linear fashion as theyemanate from an oceanic hotspot. There is generally aheavy bias in the direction of more recent colonizationtoward younger islands (Wagner and Funk, 1995),which suggests that, once ecological space is ‘‘filled’’ onthese islands, there is little further colonization.

B. Species Change Subsequent toColonization

What happens to a population subsequent to success-

ful colonization? Initially, the population may changeecologically, but over time evolutionary changes occur.

1. Ecological Change

Subsequent to the initial establishment of species innew habitats, a common outcome is ecological (orcompetitive) release, with colonizers expanding theirrange to fill the available ecological space. Regular cy-cles of distributional change following colonization of islands have been proposed several times in the liter-ature, starting with the idea suggested by E. O. Wilsonfor a ‘‘taxon cycle’’ in Melanesian ants: widespread, di-

spersive populations give rise to many more restrictedand specialized populations or species (Wilson, 1961).Although the generality of the taxon cycle has beenquestioned, the tendency for species to expand theirrange to fill the available ecological space upon initialcolonization of an area is quite predictable.

2. Evolutionary Change

As described above, the remoteness of many oceanicislands allows evolution to play a prominent role in




the addition of species to such islands, and adaptiveradiation of species from single colonizers is a frequentcharacteristic of the more remote archipelagoes. Mech-anisms underlying adaptive radiation on oceanicislands have been the subject of considerable interest

ever since Darwin described the radiation of finches inthe Galapagos (Grant, 1986). However, the extremediversity of form and modifications associated withisland colonization, coupled with the superficiallysimilar appearances of many species on differentislands within an archipelago, made it very difficultto understand how, and from where and what thediversity had arisen. The advent of molecular geneticmethods has led to a huge growth in research on oce-anic islands and has allowed the evolutionary historyof organisms to be assessed (Givnish and Sytsma,1997). Given the microcosmal nature of islands, with a

finite pool of colonizers, and the diversification of these into multiple closely related species confined tosmall areas, knowledge of the evolutionary history of the biotas has allowed some profound insights into themechanics of population divergence, and the interplaybetween isolation, sexual selection, and ecologicalshifts in allowing species formation.

Adaptive radiation, by definition, involves ecolo-gical diversification (Schluter, 2000). On oceanicislands, ecological isolation is clearly demonstrated bynumerous sympatric members of an adaptive radiation.

Within a radiation of long-jawed spiny-legged spiders

in the Hawaiian Islands, 5–8 species co-occur at a givensite, each species ecologically distinct from the otherswith which it co-occurs. Likewise, among the Hawaiianhoneycreepers, seven ecologically distinct species co-occur at a single location in high-elevation forests onthe island of Hawaii. The ecological differences amongspecies are often profound, particularly on moreremote islands where a given lineage can sometimesmonopolize the ecological arena that would be used bymultiple genera, families, or even orders, in continentalareas.

Adaptive radiation is frequently associated with an

acceleration in rate of diversification. However, the rateof diversification within an adaptive radiation may befurther enhanced by sexual selection. Recent researchhas shown that crickets in the Hawaiian Islands havethe highest documented rate of speciation known,which has been attributed to the role of courtship orsexual behavior in accelerating species divergence(Mendelson and Shaw, 2005). Likewise, sexual selec-tion is thought to have played a prominent role in thediversification of Hawaiian drosophilid fruit flies(Kaneshiro and Boake, 1987), leading to the large

number of native species (B800 in two genera, Dro-sophila and Scaptomyza) (Carson and Kaneshiro, 1976).

The initial processes of divergence leading to speciesformation on oceanic islands is still not entirely under-stood, particularly with respect to the relative impor-

tance of allopatric, parapatric, and sympatric speciation.Adaptive shifts clearly underlie species diversification inmany radiations of species in isolated islands, althoughseparation of gene pools is generally considered to be arequirement in initiating the process and has been dem-onstrated in the initial diversification of finches in theGalapagos, beetles in the Canary Islands, and spiders inthe Hawaiian Islands, among other groups. In each case,subsequent sympatry appears to have led to ecologicaldifferentiation. In the Hawaiian Islands, differentiationamong marginal isolates or founder populations appearsto have been involved in species formation within ra-

diations of Drosophila and carabid beetles. Parapatricdivergence through adaptive shifts may have played arole in the initial divergence of cave-adapted species andhas been implicated in the formation of cave endemicsfrom surface relatives in the different Hawaiian Islands(lycosid spiders and isopods) and the Canary Islands(dysderid spiders) (Roderick and Gillespie, 1998). Theidea that species on islands may be able to diverge insympatry held considerable popularity for many years,but increasing evidence from phylogenetic studies hasprovided little support to the idea. Cases where multiplesister species co-occur within a single island are most

commonly found in arthropods, and may be attributedto parapatric or ‘‘micro-allopatric’’ speciation, althoughsome authors have suggested that sympatric speciationmay have played a role in the diversification of palms onLord Howe Island (Savolainen et al., 2006) and lineageof tiny weevils (genus Miocalles) on the single smallisland of Rapa in southern French Polynesia. Together,these studies show that whatever the mechanism allo-wing initial divergence of populations, ecologicalsegregation clearly plays the primary role in allowingcoexistence of close relatives of an adaptive radiation.

C. Species Diversification

Species and the communities in which they live aredynamic entities, their existence, distribution, and abun-dance, dictated by evolutionary and ecological responsesto both natural and anthropogenic environmentalchange. For hotspot archipelagoes—those in which is-lands emanate from a single volcanic hotspot fromwhich they progressively move away—afford a uniqueopportunity: they make it possible to visualize snapshots




of evolutionary history. For example, the geological his-tory of the Hawaiian archipelago is well understood,with individual islands arranged linearly by age (Fig. 3),allowing early stages of diversification and communityformation to be studied on the island of Hawaii and

compared with progressively later stages on the olderislands of Maui, Lanai, Molokai, Oahu, and Kauai(Roderick and Gillespie, 1998). A similar chronological

arrangement is found in the archipelagoes of both theMarquesas and the Societies in French Polynesia. Be-cause of this geological history, coupled with their ex-treme isolation, the Pacific hotspot archipelagoes affordthe opportunity for phylogenetic study of community

formation within a temporal context. Other hotspotarchipelagoes, such as the Galapagos and the CanaryIslands, show temporal variation although islands in

Hawaii

160 60180160140120100

20

40

0

20

40

160 140 120 100 80 60160140120100

20

40

0

20

40

Fiji

New Zealand

Australia

Galapagos

United States

SouthAmerica

Philippines

Japan

Hawai'i

MauiMoloka'i

Lana'i

O'ahu

Kaua'i5.1 My

2.6 My3.7 My

1.9 My

1.3 My

1.3 My

0.8 My

0.43 My

0.40 My0.38 My

0.20 My

0.10 My

3.7 My

Nuku Hiva

Hiva Oa

Fatu Hiva

Tahuata

Ua Pou

Ua Huka

2.9 My

2.0 My

2.0 My

1.4 My

Marquesas

TahitiMoorea

Bora Bora

Raiatea

Tahaa

Huahine

1.0 My

1.9 My

2.3 My

2.2 My

3.0 My

3.3 My

Societies

FIGURE 3 Major hotspot archipelagoes in the Pacific. The archipelagoes of the Hawaiian Islands, Marquesas, and Society Islands are

shown on the map of the Pacific, with approximate geological ages of each island indicated. Within each archipelago, the oldest island is

to the northwest, the youngest to the southeast.




these groups are arranged as a cluster rather than a lin-ear series, yet on these islands also the evolutionaryprogression of species tends to be from older to youngerislands, and it may be possible to use the history of theseislands in the same way as that in hotspot islands, to

examine stages in adaptive radiation and communityformation.

D. Extinction

Species on islands tend to have a limited geographicdistribution and the naturally small population sizesmeans that most of these populations will not sufferadditional inbreeding depression. However, they arevulnerable to extinction as a result of demographicstochasticity. Accordingly, natural extinction rates onislands are expected to be high. In the tropical islands

of the Pacific, extinction has been much higher than onthe islands of the north Atlantic, where species rangesare much broader. The naturally small population sizesand ranges make island species more vulnerable to ex-tinction as a result of natural competition or habitatdegradation due to geological and climatic changes,and such effects have played a pivotal role in shapingisland biotas. At the same time, these same attributes of island species makes them more vulnerable to human-mediated impacts (see below).

E. Processes of Community Formation

Over the last few years, studies have started to incor-porate phylogenetic history into the investigation of community assembly (Losos, 1992). While most

of these studies argue for the nonrandom nature of community composition, they have demonstratedopposing forces in shaping communities. For exam-ple, habitat filtering in which only taxa with certaintraits colonize a community, can lead to phylogeneticclustering within that community. However, compet-itive exclusion, in which taxa that exploit similarresources will compete and exclude each other, canlead to phylogenetic overdispersion (Fig. 4). Oceanicislands have played a key role in the integration of phylogenetics and evolutionary biology into studies of community composition. Using the fact that a given

island system often includes multiple co-occurringrepresentatives of the same lineage it is possible to inferhow characters have changed to allow multiple speciesto co-occur in a community. In general, such studieshave highlighted the role of convergent evolution, andsuggest that community formation is somewhat deter-ministic. Accordingly, in a radiation of HawaiianTetragnatha spiders, ecologically and morphologicallysimilar sets of species (ecomorphs) occur in mosthabitats on all islands: one ‘‘green spiny,’’ one ‘‘large

FIGURE 4 Hypothetical scenario of phylogenetic relationships among spiders showing how different ecological forces give rise to

opposite expectations about the similarity of species traits within communities. Phylogenetic clustering may occur if closely related

species share similar physiological limitations; similar adaptations allow species with similar traits to co-occur in the same habitat.

Phylogenetic overdispersion may occur as a result of competitive exclusion among related species preventing coexistence of close

relatives that share a resource, leading to ecological divergence among close relatives.




brown spiny,’’ one ‘‘maroon spiny,’’ and one ‘‘smallbrown spiny.’’ These sets of species appear to havearrived in a habitat, either by direct colonization froman older island, or by evolution of one ecomorph fromanother on the same island, suggesting convergence

of the same set of ecological forms on each island(Gillespie, 2004).

IV. SPECIES LOSS

While the isolation of oceanic islands has made themliving laboratories for understanding species adapta-tion and evolution, it has also made them extremelyvulnerable to invasive species and other stresses. Thevulnerability of islands to extinction is now widelyrecognized. Islands have suffered much higher rates of extinction than continents (Steadman, 1995). Of ap-

proximately 100 known land bird species in the Ha-waiian Islands, almost 75% of the species present atfirst human contact have disappeared completely andmany of those that survive are in danger of extinction(James, 1995). The situation is similar for nativeplants, in which about 120 of 1000 flowering specieshave fewer than 20 individuals left in the wild, andHawaii has lost more documented arthropod speciesthan the entire continental United States. Indeed, theHawaiian Islands have been dubbed the ‘‘extinctioncapital of the world.’’ Given the information on speciesdiversification that scientists have been able to deter-

mine from oceanic islands, can we also use the islandsas tools to understand human-mediated extinction?Extinction is clearly more gradual in continents thanon many islands, but the trajectory is similar. Accord-ingly, understanding of the processes involved in spe-cies extinction in the islands, and replacement of largely native environments by species from elsewhere,may allow the development of effective conservationstrategies in mainland habitats.

A. Alien Species Invasion

One of the most severe impacts affecting many islandsystems comes from nonnative species (Sax et al.,2005). For insects, extinctions attributed to alien spe-cies invasion have been noted on Guam and theGalapagos. However, perhaps because it is the mostremote archipelago, alien species in Hawaii appear tohave taken a much greater toll on the native biota. Thenumbers now estimated for organisms purposely oraccidentally introduced into Hawaii and establishedare 3046 arthropods, 20 reptiles, 46 land birds, 19mammals, and 927 plants.

Community dynamics, and the ability of invasivespecies to penetrate certain native communities, and of certain native communities to withstand invasion, hasbeen the subject of numerous articles (Stone and Scott,1985). However, extensive research over the past few

decades has shown that multiple parameters may act(or interact) to dictate invasion success, the mostcommon being: (i) species diversity itself has beenfound, under some conditions, to serve as an ecologicalbarrier to invasion; this effect has been used to explainthe higher frequency and impact of invasions on islandswhich are considered to be relatively species poor;(ii) disturbance and the opening of ecological space canfacilitate invasion; (iii) propagule pressure, in particularthe relative abundance of native versus nonnativepropagules, can permit infiltration of alien species intonative habitats; (iv) characteristics of propagules, such

as more generalist habitat requirements, can affectinvasion success; and (v) novelty of perturbations and‘‘naıvete’’ of native biota can accentuate the impact of aninvasive species.

It is widely cited that the worst kind of introductionto an isolated archipelago is that of a generalist pred-ator or competitor capable of exploiting a broad arrayof habitats and causing secondary impacts. This cansometimes result in an effect known as ‘‘invasionalmeltdown.’’ For example, the crazy ant Anoplolepisgracilipes has invaded Christmas Island, resulting incomplete modification of the ecosystem in just 2 years.

Supercolonies of the ants extirpate the dominantnative omnivore, a large land crab, which in turnleads to increased seedling recruitment but slows litterdecomposition. At the same time, the association of the ants with plant-feeding insects increases the abun-dance of Homoptera, with associated increased im-pacts on plants including honeydew productions,which in turn enhances sooty moulds.

Devastating introductions by generalist predators/ competitors such as pigs are largely (though by nomeans completely) historical, occurring before proto-cols were established for the introduction of nonnative

species. However, abundant, small human commen-sals, such as ants (Fig. 5), mosquitoes, and agriculturaland horticultural pests, are continually expandingtheir range on to oceanic islands. Moreover, althoughthe introduction of highly specialized species for bio-control purposes has been believed to be safe in thatthey are unlikely to go beyond the confines of theirspecialized host or habitat, evidence is accumulatingin Hawaii to suggest that specialized species can ex-pand their range when faced with novel conditions,and may do so quite dramatically.




1910−20

1920−30

1930−40

1970

1978

1998

1999−2000

2005

Native range

FIGURE 5 Invasion of the little fire ant, Wasmannia auropunctata, in the Pacific. W. auropunctata is native to South and Central America. Over the last century, the species has invaded multiple

locations worldwide, and in each location its invasion has been associated with substantial impacts on the native biota. (Note that lower green represents New Caledonia and higher green the

Solomons; similarly 1999–2000 represents Fiji and Hawaii, and 2005, Tahiti.)



B. Demographic Effects—Small PopulationSizes

The limited geographic range and small populationsizes that characterize islands make them vulnerable todemographic stochasticity (see above). Species withnaturally small population sizes are similarly morevulnerable to habitat modification, simply because lossof even a small amount of habitat for a geographicallyrestricted species could easily reduce numbers belowsustainable levels.

C. Abiotic Effects—Climate Change

Islands can also be highly climate-sensitive. In the pasttwo decades, for example, short-term extreme hightemperatures contributed to a decline of coral reefsthroughout the tropics. Corals, stressed by high tem-peratures, may eject their symbiotic algae resulting incoral bleaching, which renders the corals vulnerable toany further physiological stress and many of the coloniesdie. In 1994, elevated sea temperatures killed over 90%of the living corals of American Samoa from the inter-tidal zone to a depth of 10 m and fishing catches de-clined drastically following the coral death. Likewise,during the El Nino of 1997–98, coral bleaching in Palau,known for its spectacular coral reefs, was extensive.

Other concerns of climate change relate to the factthat most of the island species have restricted distri-butions such that local vagaries of climate can elim-inate species so restricted as compared with morewidespread taxa. In particular, the high-elevationcloud forests that characterize many oceanic islandshave a narrow geographical and climatological nichethat may be eliminated with even a slight climatolo-gical change. Climate change may also precipitate de-clines in forests due to floods, droughts, or increasedincidence of pests, pathogens, or fire. In addition, in-creases in the frequency or intensity of hurricanes maycause disturbance that favors invasive species.

Sea level rise is also a concern for oceanic islands,

resulting in coastal erosion and salt water intrusion intofreshwater lenses. However, it is the low-lying islandsand atolls that are the most vulnerable from these effects.

V. THE FUTURE OF BIODIVERSITY ONOCEANIC ISLANDS

Oceanic islands have served as a source of intrigue forbiologists for centuries. Not only are they discrete

units within which the biota can be quantified andcompared, but the interplay of isolation and time hasallowed the development of uniquely derived faunason some islands. Understanding of the interplay be-tween time and isolation in allowing these unique sets

of biota to develop is still limited. Even more limited isour knowledge of the nebulous concept of ‘‘vacantniche space,’’ which appears to play a critical role infostering colonization and diversification. There ismuch to be learned from the natural patterns of diver-sification on islands. At the same time, the current rateof anthropogenically induced habitat degradation andhomogenization of the world’s fauna threatens thebiota of the world, particularly that of islands. So whatare the priorities for addressing the ongoing extinctioncrises on these islands?

A. Taxonomic Knowledge

On many islands, the taxonomic impediment is huge.Accordingly, determining the status of a species, whetherit is native or not, remains a major challenge for aconservation biologist on oceanic islands, particularlyfor those working on arthropods or other megadiversegroups. This problem is especially acute in tropical areaswhere knowledge of both potentially native and the poolof nonnative species is frequently sparse and unreliable.Research is needed to determine the identity and dis-tribution of native and alien species on different islands,

and subsequently to determine their evolutionary his-tory and assess their future trajectory.

B. Development of Realistic EvolutionaryTrajectories

In a world subject to accelerating global and localenvironmental change, policy makers and planners arestruggling to maintain the status quo of diversity innatural systems. At the same time, there is growingappreciation that biological diversity is inherently dy-

namic. Conservation strategies that do not incorporatethese evolutionary and ecological dynamics and thatignore the spatial flux in environmental quality willinevitably fail. Accordingly, an explicitly dynamic ap-proach is required to allow the assessment of the con-ditions under which areas of endemism are eithersensitive or resilient to environmental change. Oceanicislands provide circumscribed areas of discrete, highlyendemic, assemblages of organisms within which un-derstanding of evolutionary and ecological dynamicsshaping the biota is an achievable goal.

_______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y _________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________11



It is not clear what such an assessment will tell us.For example, considering the Hawaiian Islands, it isestimated that the existing angiosperm flora developedfrom 265 colonization events (not including introduc-tions or doubtfully native taxa), with the number of

species arising per colonist varying from 1 to 91. If colonization commenced on the oldest current high-elevation island of Kauai (which appears to be true formost groups), 53 successful colonizations per millionyears would be required to give rise to the current biota.In 2005, there were 1,162,000 aircraft operations (take-offs and landings) in the Hawaiian Islands, carryingapproximately 33,300,000 passengers, 500,000 t of cargo, and 130,000t of mail. If only one out of every60 passengers plus one out of 200 packages (assumingthe average package weighs 2 lb) each carried a strayseed, and just half of these seeds became established,

this would still represent an increase in the rate of col-onization to the islands of 1 billion %! Clearly the Ha-waiian Islands are no longer very isolated, and, evenwith the utmost care in preventing the introduction of new species, the ancient processes of evolution in iso-lation can never be recovered.

This does not mean, however, that the nativeenvironment on oceanic islands has ‘‘lost the game,’’nor that we should abandon the idea of long-termpreservation of the biota. It simply tells us that weneed to act quickly, and that oceanic islands canprovide much-needed insights into factors underlying

community resistance to change, and also its ability torecover completely under intensive restoration.

C. Restoration

Again because of their small size and unique biota,oceanic islands provide an opportunity to examinethe reality and practicality of restoration efforts. Forexample, dryland forests are among the most threat-ened of Hawaiian ecosystems, with one of the onlyremaining patches of the original dryland forest in

Auwahi, an area on the south slope of Haleakala onMaui. Although harboring a rich diversity of nativetree species, the area is largely pasture and dominatedby kikuya grass. However, excellent seed sources formost species still exist. So, over the last two decades,scientists have fenced 30 acres for restoration, killedthe grass, and replanted native saplings. The resultshave been quite remarkable and suggest that theentire area, complete with its arthropod fauna, can berecovered. The unique microcosmal nature of oceanicislands allows restoration efforts such as these, not

only simply to be conducted, but also to be analyzedand interpreted to allow extrapolation of the param-eters to larger and more complex continental settings.

Recognition of the acute conservation concerns im-pacting islands, and the lessons that they offer, is

becoming increasingly recognized. A recent Confer-ence of the Parties (COP8) to the Convention onBiological Diversity (CBD) placed island biodiversityhigh on the international conservation agenda.Immediate action is clearly required, not only todocument and understand the unique biota thatcharacterizes oceanic islands, but also to use thelessons learned from these relatively simple systemspotentially to address more complex global issues of conservation.

See Also the Following Articles

ADAPTATION ADAPTIVE RADIATION BIODIVERSITY,

DEFINITION OF BIODIVERSITY, EVOLUTION AND

BIODIVERSITY, ORIGIN OF BIOGEOGRAPHY, OVERVIEW

DARWIN, CHARLES (DARWINISM) DISPERSAL

BIOGEOGRAPHY DIVERSITY, COMMUNITY/ REGIONAL

LEVEL ENDEMISM EXTINCTION, CAUSES OF HOTSPOTS

INTRODUCED PLANTS, NEGATIVE EFFECTS OF ISLAND

BIOGEOGRAPHY LOSS OF BIODIVERSITY, OVERVIEW

MARKET ECONOMY AND BIODIVERSITY SPECIES-AREA

RELATIONSHIP SPECIES DIVERSITY, OVERVIEW

SPECIATION, PROCESS OF

BibliographyCarson, H. L., and Kaneshiro, K. Y. (1976). Drosophila of Hawaii:

Systematics and ecological genetics. Annu. Rev. Ecol. Syst. 7, 311–

345.

Gillespie, R. G. (2004). Community assembly through adaptive ra-

diation in Hawaiian spiders. Science 303(5656), 356–359.

Gillespie, R. G., and Roderick, G. K. (2002). Arthropods on islands:

Evolution and conservation. Annu. Rev. Entomol. 47, 595–632.

Givnish, T. J., and Sytsma, K. J. (Eds.) (1997). Molecular Evolution

and Adaptive Radiation, 621pp. Cambridge University Press,

Cambridge, UK.

Grant, P. R. (Ed.) (1986). Ecology and Evolution of Darwin’s Finches.Princeton University Press, Princeton.

Howarth, F. G., and Mull, W. P. (1992). Hawaiian Insects and Their

Kin, 160pp. University of Hawaii Press, Honolulu.

James, H. F. (1995). Prehistoric extinctions and ecological changes

on oceanic islands. In Ecological Studies (P. Vitousek, Ed.), pp.

88–102. Springer, Berlin.

Juan, C., Emerson, B., Oromi, P., and Hewitt, G. M. (2000).

Colonization and diversification: Towards a phylogeographic

synthesis for the Canary Islands. Trends Ecol. Evol. 15, 104–109.

Kaneshiro, K. Y., and Boake, C. R. B. (1987). Sexual selection and

speciation: Issues raised by Hawaiian Drosophila. Trends Ecol.

Evol. 2, 207–212.

12 _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y _______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________



Keast, A., and Miller, S. E. (Eds.) (1996). The Origin and Evolution of

Pacific Island Biotas, New Guinea to Eastern Polynesia: Patterns

and Processes, pp. 67–90. SPB Academic Publishing, Amsterdam.

Losos, J. B. (1992). The evolution of convergent structure in

Caribbean Anolis communities. Syst. Biol. 41, 403–420.

MacArthur, R. H., and Wilson, E. O. (1967). The Theory of Island

Biogeography. Princeton University Press, Princeton.

Mendelson, T. C., and Shaw, K. L. (2005). Sexual behavior: Rapid

speciation in an arthropod. Nature 433, 375–376.

Nunn, P. D. (1994). Oceanic Islands, 418pp. Blackwell, Oxford.

Roderick, G. K., and Gillespie, R. G. (1998). Speciation and

phylogeography of Hawaiian terrestrial arthropods. Mol. Ecol.

7, 519–531.

Savolainen, V., Anstett, M. C., Lexer, C., et al. (2006). Sympatric

speciation in palms on an oceanic island. Nature 441(7090),

210–213.

Sax, D. F., Stachowicz, J. J., and Gaines, S. D. (Eds.) (2005). Species

Invasions: Insights into Ecology, Evolution, and Biogeography.

Sinaur Associates, Sunderland, MA.

Schluter, D. (2000). The Ecology of Adaptive Radiation. Oxford

University Press, Oxford.

Steadman, D. W. (1995). Prehistoric extinctions of Pacific

Island birds: Biodiversity meets zooarcheology. Science 267,

1123–1131.

Stone, C. P., and Scott, J. M. (1985). Hawaii’s Terrestrial Ecosystems:

Preservation and Management, pp. 149–179. Cooperative

National Park Resources Studies Unit, University of Hawaii,

Honolulu.

Wagner, W. L., and Funk, V. A. (Eds.) (1995). Hawaiian Biogeogra-

phy: Evolution of a Hot Spot Archipelago. Smithsonian Institution

Press, Washington, DC.

Whittaker, R. J. (1998). Island Biogeography: Ecology, Evolution, and

Conservation. Oxford University Press, Oxford.

Williamson, M. H. (1981). Island Populations, 286pp. Oxford Univer-

sity Press, Oxford.

Wilson, E. O. (1961). The nature of the taxon cycle in the Mela-

nesian ant fauna. Am. Nat. 95, 169–193.

_______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y _________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________13_______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ O CEA N I C I S L A N D S : M O D EL S O F D I V ER S I T Y _________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________13

Documents

Biodiversity island models