Relevance of the Equilibrium Theory of Island Biogeography...

Relevance of the Equilibrium Theory of Island Biogeography and Species-Area Relations toConservation with a Case from AmazoniaAuthor(s): B. L. Zimmerman and R. O. BierregaardSource: Journal of Biogeography, Vol. 13, No. 2 (Mar., 1986), pp. 133-143Published by: Blackwell PublishingStable URL: http://www.jstor.org/stable/2844988Accessed: 19/08/2010 13:29

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing is collaborating with JSTOR to digitize, preserve and extend access to Journal ofBiogeography.

http://www.jstor.org

http://www.jstor.org/stable/2844988?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/action/showPublisher?publisherCode=black

Journal of Biogeography (1986) 13, 133-143

Relevance of the equilibrium theory of island biogeography and species-area relations to conservation with a case from Amazonia

B. L. ZIMMERMAN and R. 0. BIERREGAARD World Wildlife Fund-US, 1255-23rd St., N.W., Washington, D.C. 20037, U.S.A.

ABSTRACT. The equilibrium theory of island biogeography and associated species-area relations have been promoted as theoretical bases for design of nature reserves. However, the theory has not been properly validated and the practical value of biogeographic principles for conservation remains unknown. Recent studies have shown that species-area data in the absence of autecological bases provide no special insights relevant to conservation.

The unreliability of simplistic species-area data when applied to real conservation situations is illustrated with an example from the Brazilian Amazon. A prediction of area for the conservation of species of central Amazonian forest frogs was made from species-area data and found to lack relevance in light of autecological evidence.

Introduction

The equilibrium theory of island biogeography associated species-area relations have been promoted as theoretic-al bases for design of nature reserves (Terborgh, 1974, 1976; Dia- mond, 1975; Wilson & Willis, 1975; Diamond & May, 1976; Lovejoy & Oren, 1981). The actual guidelines proposed for reserve design are extracted from most theoretical papers only with some difficulty. However, the theoretical implications are that extrapolation from species-area data could predict the minimum area necessary to preserve a certain number of species.

The first section of this paper discusses generally the relevance of equilibrium theory and associated species-area relations to conservation in light of recent habitat fragment studies that emphasize ecological factors rather than just area. In the second section, the effectiveness of the species-area approach in determining a minimum sized area for preservation of 90% of the known species of

Central Amazonian forest frogs is examined and compared to an area prediction made solely from ecological data on these species.

1. The equilibrium theory of island biogeography and conservation

The number of species on an island is generally observed to increase with increasing area. The equilibrium theory of island biogeography (McArthur & Wilson, 1967) was advanced to explain this observation. The theory proposes that an island's biota is determined by a dynamic balance between the immigration of new species to the island and the extinction of species already present (McArthur & Wilson, 1967). Area sets an upper limit on the number of individuals present and the probability of any species becoming extinct will increase with smaller population sizes (Wilcox, 1980). Thejefore, species turnover rates on smaller islands will be higher. Also, the chances for successful colonization of very isolated islands

133

134 B. L. Zimmerman and R. 0. Bierregaard

are reduced and so the biotae of more isolated islands will equilibrate at lower species richness levels than those of less isolated islands. The species richness of islands can often be related mathematically to area as S=kAz, where S=species number; A=area; and k and z are constants.

Because of the similarity between islands and nature reserves, there were great hopes that the equilibrium theory of island biogeography would produce guidelines for the design of nature reserves (Diamond, 1976; Diamond & May, 1976; Terborgh, 1976; Wil- son & Willis, 1975; Lovejoy & Oren, 1981). Beyond the simple prediction that large areas will hold more species than smaller areas, it was hoped that the equilibrium theory could help to answer some of the following ques- tions: What fraction of its initial biota will a reserve eventually save and how rapidly will the remainder go extinct? How many species will survive in a reserve of a particular size? (Diamond & May, 1976). Theoretical ecologists promoted species-area relationships and models of faunal collapse derived from the equilibrium theory as analytical tools to help conservationists preserve species diversity. During the last decade there has been much mathematical modification of the theory, interpretation of relationships among parameters, and construction of species-area relationships from lists of species (see Simberloff & Abele, 1976; Pielou, 1979; and Gilbert, 1980, for reviews). Estimates from species-area regressions have even motivated specific recom- mendations about reserve size and forecast the efficiency of reserves in preserving species number following isolation (see Boecklen & Gotelli, 1984, for review). Until very recently the species-area relationship and equilibrium interpretations dominated all considerations of reserve strategy (Connor & McCoy, 1979; Reed, 1983). Indeed, the equilibrium theory of island biogeography became entrenched in conservation literature and achieved the status of a paradigm (Simberloff & Abele, 1976; Gilbert, 1980; McCoy, 1983). However, within the last 4 years there has been a call from many conservation biologists for ecologists to abandon further theoretical manipulations of island biogeographic theory and become reac- quainted with reality and the business at hand: data collection in the field (Boecklen & Gotel-

li, 1984; Lynch & Whigham, 1984; Simberloff, 1985; Kitchener et al., 1980). What happened, then, to the equilibrium theory of island biogeography and the promise of guidelines for conservation?

Apparently, after 10 years of reading island biogeography literature based on equilibrium theory, many ecologists asked themselves the same question: how exactly has my knowledge of communities and their continued function- ing within reserves been advanced? The in- escapable conclusion is that beyond the ecological truism that species richness increases with area, well known since the early nineteenth century (Connor & McCoy, 1979), the equilibrium theory of biogeography has taught us little that can be of real value planning real reserves in real places. The problem is not so much with the model itself as with the appa- rent fixation of biogeographers with the species-area relationship and their tendency to draw profound conclusions from fairly simple data, namely species lists (Abbott, 1983). The equilibrium theory deals only with numbers of species and not with their characteristics even though these may be implicit (Simberloff, 1969). Simberloff & Abele (1976) were among the first to point out that although it was apparently accepted as fact, the equilibrium theory of island biogeography had rarely been properly tested. To demonstrate the applicabil- ity of the model in any given situation it would be necessary to show a close correlation between island area and species richness, that the number of species remains constant over time, and that there is an appreciable species turnover (Margules, Higgs & Rafe, 1982). Gilbert (1980) believed that no studies had fully satis- fied these conditions. Also there is no evidence that extinction rate is area dependent (Mar- gules et al., 1982). In a scathing critique of the equilibrium theory Gilbert (1980, p. 214) says 'Some of the expositions propounded after 1967 contain some highly abstruse and some- times specious arguments judged not on empirical tests but on the difficulty of the mathema- tics or the obscurity of theoretical development'.

The equilibrium theory sparked much de- bate over whether one large reserve would preserve more species than several small reserves (Diamond, 1976; Simberloff & Abele, 1976; Terborgh, 1976; Whitcomb et al., 1976;

Equilibrium theory of island biogeography and conservation 135

Gilpin & Diamond, 1980; Higgs & Usher, 1980). It is now agreed that both equilibrium theory and the species-area relationship are equivocal on this point and cannot be used as ecological justification for the preservation of large areas over several small areas of total equivalent area or vice versa (Connor & McCoy, 1979; Higgs, 1981; Mader, 1984). Abbott (1983) showed that by using data on dispersal abilities, habitat requirements, and competitive interactions he could generate the slopes (i.e. z in S=kAz) found in island biogeographic literature. Therefore a 'z' value might reflect a variety of ecological factors rather than simply species-area relations. Con- nor & McCoy (1979), Williamson (1981), Abbott (1983) and Haila (1983) concluded that z has no unique biological meaning and therefore adds little to our understanding of species richness of islands. Another reason why the species-area relationship is not considered a reliable tool for predicting species richness is inadequacy and inconsistency of sampling method and effort. Species-area curves will not be comparable unless sampling systems are too, yet sampling method is rarely mentioned when species-area equations are discussed (Woolhouse, 1983).

The suitability of species-area regressions and models of faunal collapse for conservation practice will depend upon the quality of the predictions they generate. Boecklen & Gotelli (1984) used statistical techniques to evaluate species-area regressions and models of faunal and floral collapse from the literature. They found that the models had low explanatory power, typically accounting for only half the variation in species number. Models were sensitive to particular cases so that point estimates from species-area curves were imprecise and spanned several orders of magnitude following deletion of a single observation. Species-area models that generate imprecise estimates can support the observation that species richness increases with area but they cannot reliably determine area requirements for a given number of species (Boecklen & Gotelli, 1984).

Another objection is that the data brought to bear on the equilibrium theory have very largely been ornithological. Therefore, what little empirical evidence there is to support design strategies is really relevant only to birds

(Gilbert, 1980; Margules et al., 1982). Recent- ly, Lomolino (1984) claimed to have demonstrated that mammalian biogeography supports equilibrium theory, but his paper falls prey to the same criticisms of many of the bird papers, namely failure to consider factors other than area, failure to demonstrate species turnover, assumption of causality when only correlation had been demonstrated and no mention of survey techniques or effort. A final point: even if the equilibrium model did describe true islands, continental habitat islands or reserves are so different from oceanic islands, particu- larly regarding dispersal barriers and edge effects, that they might be badly described by it (Margules et al., 1982; Wright & Hubbell, 1983; Mader, 1984).

It is widely agreed that the equilibrium theory was accepted long before there was enough valid qualitative and quantitative evidence to support it (Connor & McCoy, 1979; Gilbert, 1980; Boecklen & Gotelli, 1984; Sim- berloff, 1985). Further focusing on details of the model is misguided because, as ecologists now agree, predictions of the model are vir- tually useless without accompanying autecological information on the species of interest. As Higgs (1981) puts it: 'Biological knowledge of the distribution and habitat requirements of the species to be conserved is necessary for the application of island biogeography and once this is available, conservation strategy can be more sensibly determined on other grounds anyway'.

The most recent island and habitat fragment biogeographic studies now consider more specific autecological factors of the species involved rather than simply area. Because these multifactor studies can more completely explain local community composition, they can also generate more precise and reliable predictions for conservation. Although they did not consider competitive effects, Graves & Gotelli (1983) showed that available habitat and main- land geographic range of species were important determinants of landbridge avifauna. Haila (1983) found that sampling effect and habitat composition strongly affected the pool of actual colonists of landbirds in a Finnish archipelago. He concluded (1983, p. 334) that 'the relation between S and A is indirect and is mediated by a chain of interrelated factors that comprise habitat composition differences in


species-specific colonization probabilities and species abundance relations'. Reed (1983) stu- died landbird populations on British islands and found species richness correlated with both distances (isolation) and area. However, there was little need for recourse to equilibrium theory-to explain these correlations because factors such as habitat composition were superior predictors (Reed, 1983). In one of the most comprehensive island biogeographic studies to date, Lynch & Whigham (1984) used point surveys to estimate the abundance and diversity of forest birds in relation to size, degree of isolation, floristics, physiognomy, and successional maturity of 270 upland forest patches in the eastern U.S.A. Structural and floristic characteristics were more important in determining species composition and local abundance of individual bird species than was patch area. The data implied that the best regional conservation strategy for some species may be to preserve the maximum total amount of breeding habitat rather than emphasizing the extent of each forest fragment (Lynch & Whigham, 1984). 'We therefore must avoid a simplistic numbers game (i.e. species lists) when we assess the biological impact of habitat disruption. Because birds differ in their ecological requirements, they also differ in their resiliency to disturbance and hence conservation importance' (Lynch & Whigham, 1984, p. 314).

II. Nature reserve design for Central Amazonian forest frogs: a comparison between species-area and ecological approaches

The purpose of this analysis is to compare prediction of nature reserve size from the equilibrium model with predictions made from autecological data and knowledge of habitat requirements for the specific animals to be preserved.

i. Species-area data

Thirty-nine species of frog were found in (strictly) primary forest of the INPA-WWF reserves near Manaus in the Central Amazon (see Lewin, 1984; Lovejoy et al., 1984) during

32 months of sampling completed in June 1985 (Table 1). Most of these species are fairly widespread throughout the Amazonian lowlands (Hodl & Zimmerman, 1986). Sampling method was identical and performed by the same principal surveyors in all reserves. We are confident that the anuran fauna of the reserves has been almost completely sampled because of the asymptotic nature of the species encounter curve. During the first 20 survey months, thirty-six species were found whereas only three species were added during the last field year. Sampling intensity was similar in all reserves in that many more nights were spent walking in the 100 and 500 ha areas than in those of 1 and 10 ha (see Zimmerman & Bogart, 1985, for sampling method details). Species lists were compiled for each reserve before any reserve was isolated from con- tiguous virgin forest by cutover (see Lewin, 1984; Lovejoy et al., 1984). Armed with species lists and corresponding survey area sizes (Table 1), we can draw a log species number versus log area graph (Fig. 1). The species- area relationships in Fig. 1 predicts than 350 ha of primary forest will include individuals of 90% (i.e. thirty-five species) of the forest frog species known in the INPA-WWF area. Pre- dictions of minimum area sizes from species- area data are not meant to be exact but may be of use at least to provide figures within the right order of magnitude. For example, it would be valuable to know whether one needs to talk in hundreds or thousands of hectares of forest when considering preservation of forest anurans. Lovejoy & Oren (1981) point out that some rare species of rainforest animals are added to species lists after observation of only

u) uj 50

o 40 ui 0- 30- U)

uO 20 0

15

m 10

D z5

z 5'

1 10 100 350500

LN AREA (HECTARES)

FIG. 1. Ln number of species versus In area (hectares) for species of central Amazonian forest frogs in seven INPA-WWF reserves (r=O.99).


TABLE 1. Species and breeding habitats of frogs observed* in numbered INPA-WWF primary forest reserves of different sizes in the central Amazon between April 1983 and June 1985.

Species Breeding 1 ha 10 ha 100 ha 500 ha habitatt 1104 1112 3209 1201 3304 1301 1401

Atelopus pulcher Str x x x Bufo dapsilis Str x x x x x Bufo marinus ? x x x x x x x Dendrophryniscus minutus Str x x Centrolenella oyampiensis Str x x x Colstethus marchesianus SPerm x x x x x x x Colostethus sp. SPerm x x x x x Pyllobates femoralis SPerm x x x Hyla boans LStr x Hyla cruentomma PerFl x x Hyla geographica Str x x x Hyla granosa Str x x x Hyla sp. LStr x Osteocephalus 'buckleyi' Treehole x x x x x x x Osteocephalus sp. ? x Osteocephalus taurinus Str, SPerm, Perfl x x x Phrynohyas coriacea ? x Phrynohyas resinifictrix Treehole x x x x Pyllomedusa bicolor Str, SPerm, Perfl x x x Phyllomedusa tarsius SPerm, Perfl x Phyllomedusa tomopterna SPerm, Perfl x x Adenomera andreae Ter x x x x x x x Eleutherodactylus fenestratus Ter x x x x x x x Eleutherodactylus sp. A Ter ? x x x x x Eleutherodactylus sp. B Ter? x Ceratophrys cornuta PerFl x x x Leptodactylus mystaceus PerFl x x x x Leptodactylus pentadactylus PerFl x x x x x x x Leptodactylus rhodomystax Str, PerFl x x x x Leptodactylus riveroi LStr x x x Leptodactylus stenodema ? x x x Leptodactylus wagneri Str x x Lithodytes lineatus ? x Chiasmocleis shudikarensis PerFl x x x Ctenophryne geayi PerFl x x Synapturanus mirandiriberoi Ter x x x Synapturanus salseri Ter x x x x x Unident. microhylid PerFl x Pipa aspera SPerm x Pipa sp. SPerm, PerFl x

Total 6 7 12 13 24 28 38 * All species listed refer to any sighting of that species and do not necessarily represent breeding populations. tStr=Stream associated; LStr=Large stream associated; SPerm=Small permanent terra firme pools;

PerFl=Periodically flooded terra firme pools; Ter=Terrestrial; ?=Breeding habitat unknown.

one or very few individuals. They suggest predicting reserve characteristics for maintain- adding a safety margin to any species-area ing breeding populations of at least thirty-five generated figure of minimum area in order to species of forest frogs? provide enough forest to encompass reasonable populations of rare species. Fig. 1 sug- .. gests that about 500 ha would include popula- i tions of 90% of the forest interior frog species Almost 3 years fieldwork in the INPA- in the Central Amazon. But how do species- WWF reserves has shown that a critical factor area data fare compared to ecological data in limiting distribution of many and probably


most species of forest frog is breeding habitat. Eleven species require landlocked, permanent or periodically flooded terra firme (i.e. upland and not stream valley) pools for breeding and some of these species are even particular about whether the pool is muddy (as in wallows of the collared peccary, Tayassu pecari) or clear water (as are the periodically flooded sites). Nine species appear somewhat less restricted by breeding habitat as they will breed in either terra firme or streamside rainpools. Seven species are strictly streambreeders including three that require large streams greater than 3 oi' 4m wide. Five terrestrial breeding species are apparently unrestricted by breeding habitat; two are treehole breeders; and nothing is known about breeding biology of eight species (Table 1). Therefore at least one half of the forest frogs require specialized habitat other than streams themselves for reproduction. Be- sides streams, the conspicuously important habitats for forest anuran reproduction in the Central Amazon are small, permanent terra firme pools (often Tayassu pecari wallows); sporadically flooded terra firme pools; and the more extensive floodplains of large streams where many small pools form during rains.

iii. Comparison of species-area and autecological approaches

Consider the terra firme peccary wallows and other small permanent pools in which eleven species breed, including four that repro- duce exclusively at such sites. As of July 1985, three sites with terra firme permanent pools were known along 88km of INPA-WWF trail. Because a peccary wallow or other terra firme permanent pool can be detected for about 10 m on either side of the trail (a very conservative estimate), we estimate a density of nine small, permanent pool sites per 500 ha. Similarly, we know of eleven periodically flooding sites (exclusively used for reproduction by at least seven species; Table 1) along 88 km of trail and these therefore occur at a density of sixteen per 500 ha. Also, a large stream is found only in a small corner of reserve 1401. If these sites are independently distributed throughout forest, then we expect to include nine permanent, pig wallow-type pools, a section of large stream, and sixteen periodically flooded sites within a randomly chosen 500 ha plot. This

number of breeding sites might easily be enough to maintain populations of the relevant species and we have no evidence to the con- trary. However, there are ecological factors that will significantly decrease the ability of a randomly chosen 500ha to preserve breeding populations of 90% of the forest frogs.

First, periodically flooded pools vary con- siderably in quality with respect to breeding by frogs. The largest three ponds in our area support breeding populations of most of the relevant species (Table 1) that are at least 3 times as large as those observed at eight smaller sites. Also, only the same four of eleven periodically flooded pools flood every year. Two of the periodically flooded areas both flood annually and are large (i.e. greater than 5 x 5 m) and therefore must produce many more new frogs than the other nine sites. Independently distributed and occurring at a density of two per 88km, we would expect to include only three high quality breeding habitat sites within a randomly chosen 500ha. However, the most important factor that

militates against the effectiveness of a randomly chosen 500ha area in preserving an intact forest anuran community is the clumped, rather than independent distribution of the periodically flooded sites and the rarity of large streams. Sites were classified as small, permanent, 'pig wallow-type' terra firme pools; periodically flooded terra firme pools; or large stream and plotted as they are encoun- tered along an 88 km 'supertrail' running through the survey area from east to west (Fig. 2). The observed distribution of these sites along the 'supertrail' was then compared to the expected if sites were uniformly randomly distributed throughout forest. The supertrail was divided into eight llkm sections and the Poisson distribution used to calculate the expected number of trail segments with none, one, two, and three or more pools. For the temporarily flooded sites ('F' in Fig. 2), which occur at a mean density of 0.125/km, the value of the chi-square statistic comparing observed and expected values, X2=7.835, 3 d.f., P<0.05 (see Table 2).

Periodically flooded sites are spatially clumped whereas small, permanent pools are independently distributed throughout the forest (see Table 2). Large streams are rare in the INPA-WWF area with only one occurring


F FF FF S S S

0 KMS 10 20 30

FF F FF F P P

40 50 60

F

70 80 88

FIG. 2. Consecutive kilometres of trail surveyed for anurans in INPA-WWF reserves between April 1983 and May 1985 with positions of breeding sites indicated. F=periodically flooded terra firme pool; P=small, permanent, 'pig wallow'-type terra firme pool; S=large stream.

in a small part of some 3500 ha of reserve. This result means that although we can probably count on including several, 'pig wallow-type' pools in a randomly chosen 500ha, there is no guarantee of including any periodic pool sites or a large stream, much less the highest quality breeding pools or the simultaneous occurrence of periodically flooded pools and a large stream.

The importance of having all three breeding habitats in one reserve should be emphasized. The 'pig wallow-type' pools are undoubtedly critical to reproduction of some species such as Colostethus marchesianus, Colostethus sp., Phyllobates femoralis and Pipa sp., but they support only small breeding populations of other species such as Phyllomedusa tarsius,

TABLE 2. Significance of clumped distribution of frog breeding pools in the INPA-WWF reserves

For temporarily flooded sites:

Expected Observed No. of trail segments with:

0 pools 2.0227 5 1 pool 2.2781 1 2 pools 1.9121 0 -3 pools 1.2840 2

,X2=7.835, P<O.05

For pig-wallows: Expected Observed

No. of trail segments with: 0 pools 5.4978 5 1 pool 2.0622 3 -2 pools 0.4400 0

,X2=0.912, NS

Phyllomedusa bicolor, Phyllomedusa tomopterna and Osteocephalus taurinus which also breed at periodically but annually flooded sites. Since large, periodically but annually flooded pools are relatively rare in forest, the small, permanent pools may act as stepping stones for the population to disperse among the larger breeding sites. Reproduction by forest frogs also concentrates along large streams that have extensive, pool-forming floodplain. Nine species breed strictly at streams and three are associated only with large streams. Breeding populations of all the stream species are always many times larger at large stream habitat than along equal lengths of small streams. Therefore the conservation area that will support the greatest number of breeding individuals of the maximum number of forest frog species should include periodically flooded, terra firme pools with at least one large site that floods annually; several small, permanent 'pig wallow-type' pools; and a section of large stream habitat.

With some understanding of breeding habitat requirements for each species and without recourse to species-area data, we can choose high quality areas with certainty and thereby maximize chances that a conservation area preserves species. 100 ha containing quality breeding habitat would be of far greater conservation value than 500ha containing little, subquality, or even no critical habitat types. We can illustrate this situation with a real example from the INPA-WWF area. Two 500 ha areas were surveyed with respect to habitat. (One area is part of reserve 1401, Figs. 1 and 2, Table 1, and the other, 3402; reserve 3402 was not included in analysis of


species--area data because sampling effort with respect to frogs was not equal to that performed at 1401.) One 4 km section out of 16km of trail surveyed in 1401 passes by four periodically flooded sites of which two are large and flood annually, one permanent 'pig wallow-type' pool, and a section of large stream (Fig. 2). In contrast, habitat surveys in 3402 found only two pig wallows and no periodically flooded sites or large stream even though more than 30 km are walked in this reserve.

The uneven distribution of frog breeding habitat in forest must be considered when interpreting Fig. 1. Without reflecting on ecological data and sampling biases, one might take the remarkably smooth linear relationship exhibited by Fig. 1 as evidence for Preston's (1960, 1962) and MacArthur & Wilson's (1967) 'area per se' (as Connor & McCoy (1979) called it) interpretation of species-area data. The 'area per se' hypothesis explains species number as a function of immigration and extinction rates (Connor & McCoy, 1979). However, we know that at least six strictly stream associated species were observed in the two 100 ha reserves and single 500 ha area (Table 1) because these reserves have streams whereas the smaller reserves do not. Likewise, in the 500 ha of reserve 1401, at least eight species were undoubtedly observed during the survey period because this reserve has permanent pools and extensive stream floodplain that the others do not. Also in reserve 1401, these specialized breeding habitats happen to be concentrated within about 150 ha of the 500 and if we plotted this 150 ha area on Fig. 1 the point would lie well above the regression line. Similarly, eleven species were observed in two thoroughly sampled 10 ha reserves (Fig. 1, Table 1) that contain no longlasting water on the ground, whereas nineteen species have already been recorded in preliminary surveys of a third 10 ha reserve with a semipermanent pond. Observation of some twenty-one species of INPA-WWF forest frog is correlated with the presence of specific habitat types, and of course, the larger the area surveyed, the greater the chance of including these habitats. These data essentially support a habitat heterogeneity interpretation of species- area relations (Williams, 1964; see Connor & McCoy, 1979, Williamson, 1981, and Simber- loff, 1985, for reviews); there is no evidence that immigration and extinction rates play a role.

Another factor complicates interpretation of Fig. 1. That is, all species of forest frogs with the exception of six strictly streamside inhabi- tants, can be found distant from their respec- tive breeding habitats in any part of the forest. Also, some species in any one area breed on only two or three nights a year or even not at all because breeding pools do not necessarily flood annually. Therefore, individuals of many species spend much time not associated with their breeding grounds. Given that nonbreed- ing frogs do occur generally throughout the forest, 500ha derived from Fig. 1 may indeed be a good estimation of the minimum area necessary to include reasonable population sizes of 90% of the forest species. But without sufficient breeding habitat, no area can preserve species of forest frogs.

iv. Reserve design for Central Amazonian forest frogs: conclusions

In light of ecological information on the species involved, emphasis on species-area guidelines for conservation of forest frogs is misfounded. Critical breeding habitat must be identified and places that contain quality habitat in high density found before the reserve size question is addressed. This is not to say that the minimum size question is unimpor- tant, because ultimately we must preserve enough area to ensure the maintenance of viable population sizes (Schaffer, 1981). Once critical breeding habitats are found, they should be assessed by mark and recapture studies to see how many individuals of each species breed there, and therefore roughly how many of each habitat type are necessary in an area to sustain the species. Also, in the INPA- WWF area there is evidence that the collared peccary (Tayassu pecari) may increase breeding habitat available for several forest frog species by keeping open or enlarging permanent terra firme pools in which the pigs dig and cavort. It may be that the minimum critical size of area to preserve a forest frog fauna differs depending on the presence or absence of Tayassu pecarn.

The question arises as to who would ever consider making a reserve of a few hundred hectares for Amazonian frogs anyway? After all, national parks deal in tens and hundreds of thousands of hectares, which is surely more


than enough area to include the pools, ponds and streams required by frogs. However, in the Amazon as everywhere else, the forest is being rendered into isolated fragments by civilization. In Brazil where, by law, 50% of any tract of land scheduled for development must be left as forest, it is often of no importance to developers where the surviving pieces are left. With conservation guidelines available for each animal group, these remnant areas could be chosen to maximize breeding habitat and other critical resources for animals.

Conclusions

A widespread opinion among ecologists today is that predictions derived from the equilibrium theory of island biogeography and species-area relations concerning design of nature refuges have been neither helpful or warranted and that there now exists an urgent need for autecological data on the species to be preserved (Simberloff & Abele, 1976, 1982; Whitcomb et al., 1976; Connor & McCoy, 1979; Gilbert, 1980; Wilcox, 1980; Higgs, 1981; Humpreys & Kitchener, 1982; Margules et al., 1982; McCoy, 1982, 1983; Abbott, 1983; Graves & Gotelli, 1983; Haila, 1983; Wool- house, 1983; Boecklen & Gotelli, 1984; Lynch & Whigham, 1984; Mader, 1984; Simberloff, 1985). It is true, as Diamond (1984) points out, that the rapid pace of environmental destruc- tion will preclude the feasibility of thorough autecological studies in many ecosystems to be preserved. Diamond (1984) maintains that in such ecologically sightless situations, species- area data can act like a blindman's cane and show us roughly the way. Our results and those of other empirical studies, however, lead us to conclude that calculation of reserve sizes based solely on species-area data can never be more than uninspired guessing. Intuitive guessing about characteristics of a faunal reserve made by the field biologists involved would probably achieve better conservation results. If the impressive brainpower and effort used in repeated vain attempts to extract conservation strategy from biogeographic theory were in- stead devoted to autecological research, how much better would conservation be served?

Theoretical models do have a place in ecology but as means to various ends rather than as

ends in themselves (Pielou, 1981; Schaffer, 1981). It seems that theoretical ecologists often construct, refine and elaborate models with little effort to link them with the real world through the use of empirical data (Pielou, 1981). Simberloff (1983, p. 629) asks, 'How many of the pages on evolutionary ecology theory in, for example, The American Natural- ist, are misguidedly focused on theory and how many are acually required to present the model that motivates tests of hypotheses?' Responding to Roughgarden's (1983) charge that the main reason for criticism of theory is because of naive and usually incorrect popular impressions, Simberloff (1983, p. 629) says, 'I presume by popular impressions, Roughgarden (1983) refers to interpretations by us benighted non-mathematicians and by the non-ecologists who have to determine what practical lessons ecology yields ... The bloated theoretical literature has consequences beyond being a chore for ecologists to wade through'. We would suggest that a decade of island biogeography theory is a case in point.

We field biologists and conservationists do not need another paper discussing species lists and species-area relationships or z values; we urgently need concrete statements about the biology of individuals animal and plant species. There are no shortcuts; the unique relationships of each ecosystem must be under- stood before we can save them and this will come only with time spent in the field.

Acknowledgments

This study is part of the Minimum Critical Size of Ecosystems project; a binational research effort towards understanding ecology of Cen- tral Amazonian forest by the Instituto Nacion- al de Pesquisas da Amazonia (INPA) and World Wildlife Fund-US. We are also very grateful to World Wildlife Fund-Canada and the George D. Harris Foundation for funds they provided. Manuel Rego, Ocirio Perreira, Laercio Reis, Paulo Vanzolini and Tom Love- joy contributed indispensable aid.

The study represents publication number 40 in the Minimum Critical Size of Ecosystems Project (Dinamica Biologica de Fragmentos Florestais) Technical Series.


References

Abbott, I. (1983) The meaning of z in species-area regressions and the study of species turnover in island biogeography. Oikos, 41, 385-390.

Boecklen, W.J. & Gotelli, N.J. (1984) Island biogeographic theory and conservation practice: species-area or specious-area relationship? Biol. Conserv. 29, 63-80.

Connor, E.F. & McCoy, E.D. (1979) The statistics and biology of the species-area relationship. Amer. Natur. 113, 791-833.

Diamond, J. (1975) The island dilemma: lessons of modern biogeography studies for the design of nature preserves. Biol. Conserv. 7, 129-146.

Diamond, J. (1976) Island biogeography and conservation: strategy and limitations. Science, 193, 1027-1029.

Diamond, J. (1984) Distributions of New Zealand birds on real and virtual islands. N.Z. J. Ecol. 7, 37-55.

Diamond, R.M. & May, R.M. (1976) Island biogeography and the design of natural reserves. Theoretical ecology (ed. by R. M. May), pp. 163-186. Saunders, Philadelphia.

Gilbert, F.S. (1980) The equilibrium theory of island biogeography: fact or fiction? J. Biogeogr. 7, 209-235.

Gilpin, M.E. & Diamond, J.M. (1980) Subdivision of nature reserves and the maintenance of species diversity. Nature, 285, 567-568.

Graves, G.R. & Gotelli, N.J. (1983) Neotropical landbridge avifaunas: new approaches to null hypotheses in biogeography. Oikos, 41, 322-333.

Haila, Y. (1983) Land birds on northern islands: a sampling metaphor for insular colonization. Oikos, 41, 334-351.

Higgs, A.J. (1981) Island biogeography theory and nature reserve design. J. Biogeogr. 8, 117-124.

Higgs, A.J. & Usher, M.B. (1980) Should nature reserves be large or small? Nature, 285, 568-569.

Hodl, W. & Zimmerman, B. (1986) Frogs of the Amazonian Lowlands. (In preparation.)

Humphreys, W.F. & Kitchener, D.J. (1982) The effect of habitat utilization on species-area curves: implications for optimal reserve area. J. Biogeogr. 9, 391-396.

Kitchener, D.J., Chapman, A., Dell, J., Muir, B.G. & Palmer, M. (1980) Lizard assemblage and reserve size and structure in the western Austra- lian wheatbelt-some implications for conservation. Biol. Conserv. 17, 25-62.

Lewin, R. (1984) Parks: how big is big enough? Science, 225, 611-612.

Lomolino, M.V. (1984) Mammalian island biogeography: effects of area, isolation, and vagility. Oecologia, 61, 376-382.

Lovejoy, T.E. & Oren, D.C. (1981) Minimum critical size of ecosystems. Forest island dynamics in man dominated landscapes (ed. by R. L. Burgess and D. M. Sharp), pp. 7-12. Springer, Berlin.

Lovejoy, T.E., Rankin, J.M., Bierregaard, R.O., Brown, K.S., Jr, Emmons, L. & VanderVoort,

M.E. (1984) Ecosystem decay of Amazon forest remnants. Extinctions (ed. by M. H. Nitecki), pp. 295-325. University of Chicago Press.

Lynch, J.F. & Whigham, D.F. (1984) Effects of forest fragmentation on breeding bird communities in Maryland, U.S.A. Biol. Conserv. 28, 287-324.

MacArthur, R.H. & Wilson, E.O. (1967) The theory of island biogeography. Princeton University Press.

Mader, H.J. (1984) Animal habitat isolation by roads and agricultural fields. Biol. Conserv. 29, 81-96.

Margules, C., Higgs, A.J. & Rafe, R.W. (1982) Modern biogeographic theory: are there any lessons for nature reserve design? Biol. Conserv, 24, 115-128.

McCoy, E.D. (1982) The application of island biogeography to forest tracts: problems in deter- mination of turnover rates. Biol. Conserv. 22, 217-227.

McCoy E.D. (1983) The application of island biogeographic theory to patches of habitat: how much land is enough? Biol. Conserv. 25, 53-61.

Pielou, E.C. (1979) Biogeography. Wiley, New York.

Pielou, E.C. (1981) The usefulness of ecological models: a stock taking. Q. Rev. Biol. 56, 17-31.

Preston, E.W. (1960) Time and space and the variation of species. Ecology, 41, 611-627.

Preston, E.W. (1962) The canonical distribution of commoness and rarity. Ecology, 43, 185-215 and 410-432.

Reed, T.M. (1983) The role of species-area relationships in reserve choice: a British example. Biol. Conserv. 25, 263-271.

Roughgarden, J. (1983) Competition and theory in community ecology. Amer. Nat. 122, 153-601.

Schaffer, M.L. (1981) Minimum population sizes for species conservation. Bioscience, 31, 131-13,4.

Simberloff, D. (1969) Experimental zoogeography of islands: a model for insular colonizations. Ecolo- gy, 50, 296-314.

Simberloff, D. (1983) Competition theory, hypothesis testing and other community ecological buzz- words. Amer. Nat. 122, 626-635.

Simberloff, D. (1985) Design of nature reserves. In: Wildlife conservation (ed. by M. B. Usher). Chapman & Hall, London.

Simberloff, D.S. & Abele, L.G. (1976) Island biogeographic theory and conservation practice. Science, 191, 285-286.

Simberloff, D. & Abele, L.G. (1982) Refuge design and island biogeographic theory: effects of fragmentation. Amer. Nat. 120, 41-50.

Terborgh, J. (1974) Preservation of natural diversity: the problem of extinction prone species. Biosci- ence, 24, 715-722.

Terborgh, J. (1976) Island biogeography and conservation: strategy and limitations. Science, 193, 1029-1030.

Whitcomb, R.F., Lynch, J.F., Opler, P.A. & Rob- bins, C.S. (1976) (Response to Simberloff & Abele, 1976). Science, 193, 1031-1032.

Equilibrium theory of island biogeographv and conservation 143

Wilcox, B.A. (1980) Insular ecology and conservation. Conservation biology: and evolutionary perspective (ed. by M. Soule and B. Wilcox), pp. 95-117. Sinauer Associates, Massachusetts.

Williamson, M. (1981) Island populations. Oxford University Press.

Wilson, E.O. & Willis, E. (1975) Applied biogeography. Ecology and evolution of communities (ed. by M. L. Cody and J. M. Di- amond), pp. 522-534. Harvard University Press.

Woolhouse, M.E.J. (1983) The theory and practice

of the species-area effect applied to the breeding birds of British woods. Biol. Conserv. 27. 315- 332.

Wright, S.J. & Hubbell, S.P. (1983) Stochastic extinction and reserve size: a focal species approach. Oikos, 41, 466-476.

Zimmermann, B.L. & Bogart, J.P. (1985) A structural and behavioural analysis of vocalization by central Amazonian forest frogs. Acta Amazonica. (in press).

Relevance of the Equilibrium Theory of Island Biogeography...

Documents

Transcript of Relevance of the Equilibrium Theory of Island Biogeography...