Preliminary Assessment of the Flagship Species Concept at a Small Scale

of 8

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
8 pages
0 downs
Artículo científico
   Animal Conservation  (2004)  7 , 63–70  C  2004 The Zoological Society of London. Printed in the United Kingdom DOI:10.1017/S136794300300115X Preliminary assessment of the flagship species conceptat a small scale Tim Caro 1 , 2 † , Andrew Engilis Jr  1 , 2 , Emily Fitzherbert 1 , 3 and Toby Gardner  1 , 3 1 Las Cuevas Research Station, P.O. Box 410, Belmopan, Belize 2 Department of Wildlife, Fish and Conservation Biology, University of California, Davis, CA 95616, USA 3 Centre for Ecology, Evolution and Conservation, School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, UK  (Received 26 February 2003; accepted 1 August 2003) Abstract Flagship species are charismatic species that serve as a symbol and rallying point to stimulate conservationawareness and action but are often used synonymously as  de facto  umbrella species to delineate reserve boundaries. We tested the extent to which the presence of a flagship species would protect other ‘background species’ at the local scale at which practical decisions about small reserves are often made. Using long termsightings, we identified four 1 km 2 sites that are frequently visited by jaguars and by tapirs (flagship species),and by white-lipped peccaries and spider monkeys (non-flagship species) in neotropical rainforest in Belize.We then made inventories of five vertebrate taxonomic groups at each site. We found no consistent differencesin species richness or abundances of frogs, phyllostomid bats, terrestrial mammals, scansorial mammals or  birds across the four sites, except that frog diversity and abundance was higher close to a river at the flagshipsitewheretapirswerefound.SincetheseclassicLatinAmericanflagshipspeciesfailtoencompassparticularlyhigh numbers or abundances of vertebrate species at a local scale, they appear to be a poor conservation toolwhen co-opted as umbrella species for delineating the location of very small reserves in the neotropics. INTRODUCTION While reserves are an integral part of any conservationeffort (Margules & Pressey, 2000), the window of opportunity that we have for setting aside new reservesis limited given current human population growthand migration into relatively pristine areas. Sincewe lack fundamental data on species richness and abundance in most areas, scientists have resorted tousing surrogate species to facilitate the establishment of reserves (Simberloff, 1998; Caro & O’Doherty, 1999).For example, at a large geographical scale attempts have been made to test whether the distribution of one well-knowntaxonomicgroup(abiodiversityindicator)predictsthe distribution of other, lesser known taxa (e.g. Ricketts etal  ., 1999). Results have been mixed: at a continental or nationalscale,patternsofspeciesrichnessofdifferenttaxasometimescoincide(e.g.Pearson&Cassola,1992;Myers et al  ., 2000) but not always (Prendergast  et al  ., 1993;Dobson  et al  ., 1997; Kerr, 1997), whereas, at a smaller,county-wide or regional scale, where decisions aboutreserves are usually made, hotspots of species richnessin one group are usually poor predictors of those in other groups (Flather   et al  ., 1997; Lawton  et al  ., 1998; Reid,1998; van Jaarsveld   et al  ., 1998; Andelman & Fagan, † All correspondence to: T. Caro. Tel: 530-752-0596; Fax: 530-752-4154; E-mail: 2000; Chase  et al  ., 2000). At present, there is intenseinterest in biodiversity indicators in both academic and applied circles (e.g. Williams  et al  ., 1996; Pressey  et al  .,1999; Virolainen  et al  ., 1999; Lindenmayer, Margules& Botkin, 2000; Rodrigues, Gregory & Gaston, 2000) but their practical utility in establishing the location of reserves has yet to be demonstrated.Umbrella species have also been employed as tools for delineating the boundaries of reserves but at a smaller scale. Umbrella species are ‘species with large arearequirements, which if given sufficient protected habitatarea, will bring many other species under protection’(Noss, 1990; Caro, 2003). Tests of whether umbrellaspecies do offer protection for ‘background species’suggest that they can be guardians of species with similar ecology to the umbrella species (Martikainen, Kaila &Haila, 1998; Swengel & Swengel, 1999) or guardians of species on which the umbrella species depends (Launer&Murphy, 1994). Nevertheless, there are virtually no testsof whether umbrella species adequately protect large populations of background species (but see Berger, 1997;Caro, 2003).Finally, flagship species, the focus of this paper, areused in a strategic role to raise public awareness, actionand funding (Western, 1987; Leader-Williams & Dublin,2000). A flagship species has been variously defined as(1) a popular charismatic species that serves as a symboland rallying point to stimulate conservation awareness  64 T. C ARO  ET AL. and action (Heywood, 1995), (2) a species that drawsfinancialsupportmoreeasily. ... andbysodoingservesto protect habitat and other species under the ‘umbrella’ of their large habitat requirements (Meffe & Carroll, 1997),(3) a species that has become a symbol and leadingelement of an entire ecosystem campaign (Simberloff,1998) and (4) normally a charismatic large vertebrate thatcan be used to anchor a conservation campaign because itarouses public interest and sympathy (Simberloff, 1998).Flagship species have been used to raise funds and  public awareness globally, for example, the giant panda(  Ailuropoda melanoleuca ) symbol of the World WideFundforNature(Caro&O’Doherty,1999);forlargeareasatthescaleofanecoregion,forexamplejaguars(  Pantheraonca ) in Latin America (Sanderson  et al  ., 2002 a ); and atsmaller scales when conservation organisations want tohelp to establish reserves, for instance, the CockscombBasin Wildlife Sanctuary in Belize (Rabinowitz, 1986).Apartfromtheirstrategicuse,flagshipspeciesareoftenmisappliedorusedtofulfiltheroleofumbrellaspecies(for a review, see Leader-Williams & Dublin, 2000), indeed the Meffe & Carroll definition states this explicitly. Theeliding of these two concepts occurs at three scales. At alarge scale, the geographical range of a flagship speciesoften encompasses the geographical ranges of many other species (although this has never been shown explicitly);at an intermediate scale, its presence may predict the presence of other species although this is equivocal(Andelman & Fagan, 2000); and at a small, ecologicalscale,flagshipspeciesareusedtoestablishsmallreserves.As an illustration of the last point, local people noticethat individuals of a charismatic species aggregate ina particular area, its location becomes well known, theinformation is then passed on to visitors from the city, it issubsequently discussed by officials and non-governmentorganisations (NGOs), and eventually incorporated intoa conservation plan. Small reserves established usingflagship species include Pelican Island National WildlifeRefuge, Florida, USA; Avenue of Giants Redwood StateandNationalParks,California,USA;ElRosarioMonarchButterfly Sanctuary, Michoacan, Mexico; Lake Nakuru National Park, Kenya for flamingoes (  Phoenicopterusruber   and   P. minor  ); and Ujung Kulon National Park,Indonesia for Javan rhinoceroses (  Rhinoceros sondaicus )and Javan tigers (  Panthera tigris ). Unfortunately, wedo not know whether such an  ad hoc  procedure of conflating flagship and umbrella attributes to a speciesis an effective conservation strategy (Pressey, 1994).We therefore decided to develop a method and makea preliminary assessment of the conservation utility of flagship species in a neotropical rainforest context. To dothis, we examined biodiversity in a limited area regularlyvisited by a flagship species because small nature reservesin the neotropics are often set up in these very areas.For example, in Belize, where this study was conducted,Guanacaste National Park and Tapir Mountain NatureReserve are only 50 and 6741 acres, respectively, and each was set up using flagship species.We chose two species of mammal that have been used as flagship species in neotropical rainforest and two thathave not. Our first species was the jaguar, which has beenused to raise conservation awareness throughout muchof Latin America and was instrumental in attracting the political attention necessary to set up the CockscombBasin Wildlife Sanctuary in Belize. It is a wide rangingcarnivore that has an average male range of 33.4 km 2 (Rabinowitz, 1986; Rabinowitz & Nottingham, 1986) thatislikelytoencompasspopulationsofmanysmallerspeciesand therefore conforms to a classic umbrella species. Our secondflagshipspecieswasBaird’stapir( Tapirusbairdii ),the national mammal of Belize and a species used insetting up the Tapir Mountain Nature Reserve there (seealso Downer, 1996). It has a relatively small home range(1.8km 2 ; Williams, 1984) and favours riparian habitat(Fragoso, 1991). Our third species was the white-lipped  peccary (  Dicotyles pecari ) known to travel in groups over large200km 2 areas(Kiltie&Terborgh,1983).Itiswidelyregarded as a source of food rather than an object of aweor affection by Latin Americans and is not therefore aflagshipspecies.Ourfourthspecieswasthespidermonkey(  Ateles geoffroyi ), a locally common species with a smallhome range (1.1 km 2 : Rowe, 1996) and a source of meatin many areas of south and central America, although notin Belize, again not a flagship species.Our approach focuses on the way in which the locationof reserves are commonly decided in the real world  because we targeted our investigation specifically to thoseareas where people see surrogate species in the course of their day-to-day activities rather than using the populationrange of a flagship species, which is unknown in thevast majority of cases. If vertebrate species richnessand abundance are higher in foci of tapir and jaguar activity, it would indicate that flagship species are a usefulconservation tool at a small, local scale in the neotropics;ifnodifferences arefound between sitesitwouldquestionthe utility of using flagship species as a  de facto  umbrellaspecies at a small scale. METHODS The study was conducted in the centre of ChiquibulForest Reserve, western Belize, near the Las CuevasResearch Station (16 ◦ 44   N, 88 ◦ 59  W, elevation 500 m).Theareaconsistsofamosaicofdeciduoussemi-evergreenand deciduous seasonal forest receiving an average of 1500–2000 mm of rain per annum (Wright  et al  ., 1959;Hartshorn  et al  ., 1984). The study was carried out in Julyand August 2001 at the start of the wet season. Centresof activity of jaguars, tapirs, white-lipped peccariesand spider monkeys were determined from the stationsightings book, which has been kept up by visitingscientists and station personnel continuously from August1996 to the time of the study. Specifically, we plotted thelocation of every sighting of these four species on a mapdivided into 1 km 2 grids (Fig. 1). For tapir (Fig. 1(c)), wechose the grid square with the most sightings; for jaguar (Fig.1(a))andwhite-lippedpeccary(Fig.1(b)),wepicked a square with a large number of sightings that was mostremote from Las Cuevas Research Station since informal  Flagship species 65 (a) (b)(c) (d) Fig. 1.  Location of sightings of (a) jaguar, (b) white-lipped peccary, (c) Baird’s tapir, and (d) spider monkey taken from the Las CuevasResearch Station sightings book between August 1996 and July 2001. Numbers refer to the total number of sightings in each 1 km × 1 kmgrid square; x denotes the location of the Research Station; the continuous line is the Macal River; dashed lines show gravel roads; dotted and dashed lines mark minor tracks that require clearing each year. Grid squares outlined in bold show the square in which biologicalinventories were conducted. Note that a jaguar was seen in the bold square in (a) during one of the days that it was surveyed, white-lipped  peccaries were heard in the bold square in (b) on 2 of the days that it was being surveyed and spider monkeys were seen in the bold squarein (d) on 2 of the days that it was being surveyed. sightings were more likely to be made near the station by chance. For spider monkeys (Fig. 1(d)), we chose thesquare with one less sighting than the maximum so as to be able to use vehicle transport to carry mammal trapsto the site. We then restricted our biological inventory tothese four 1 km 2 sites and compared vertebrate speciesrichness and abundance between them.The number of ground dwelling frog species and frogabundance were recorded by walking permanent transectsof known length in each of the four sites. Transect lineswere stratified around areas presumed to be of both highspecies diversity (creeks and rivers) and areas noted to be ecologically distinct in terms of habitat. Each line wascut to be 500 m in length and 2 m wide. Transect lineswere surveyed by two people who walked down the lineeach disturbing leaf litter or branches on one half of thetransect. Search effort was standardised by limiting it to1 h per transect (0.5 km/h). All transects were surveyed during the day and night, except for the transect by theriverwhereearlyseasonfloodingpreventedcompletionof the same number of transects as other sites; to minimisethe effects of disturbance, transects were never revisited within 36 h. Frogs were counted and identified to specieslevel using Campbell (1998) and Lee (2000). The totalnumber of species in Chiquibul ( n = 21 species includingspecies seen in trees) was calculated by combining thisstudy with a survey conducted by E.F. and T.G. in 2000(Gardner & Fitzherbert, 2002).To determine species richness and abundance of  phyllostomid bats, two 2 m × 10 m mist nets were setin active flyways determined by the presence of water (streams or forest ponds) or gaps in the forest. Each netwas opened prior to sunset and left open through thenight. Nets were checked each morning. Bats were either released or collected as voucher specimens and the netswere then closed for the day. Bats were identified in thefield using primarily Reid (1997) and Emmons (1990).Specimens were deposited and determinations completed at the Museum of Wildlife and Fish Biology, Universityof California, Davis using Hall (1981). Bat abundancewas expressed as the number of individuals per night. Thenumber of phyllostomids ( n = 28) in the Maya Mountains(which contains the Chiqibul Forest Reserve) was takenfrom McCarthy (1998).To determine richness and abundance of smallmammalsweusednodaltrapping(trapssaturatedarounda pointintheforest)alongthreeseparatetransectsthatwerewidely separated within the 1 km 2 grid square. The nodes(10/transect) were selected using animal sign, diversityof forest floor structure and liana growth. Along eachtransect we set 30 ‘Victor’ or ‘Museum Special’ snaptraps, 40 Sherman (23 × 8 × 8 cm) traps and 15 similar   66 T. C ARO  ET AL. custom-made wire mesh traps as described by Emmons(1984). To sample medium-sized mammals, we set 10Tomahawk (40 × 13 × 13 cm or 40 × 17 × 17cm) trapsand we set five large Tomahawk (65 × 22.5 × 22.5 cm)traps to catch larger terrestrial mammals. In each of our four sites, we set this combination and number of traps(100) along a stream, river or beside a pond, 100 withinthe forest interior and 100 along the forest edge, usuallya trail (i.e. 300 traps/site in total). Snap traps and somesmall live traps were tied to trees and vines, 1–3 m aboveground; others were placed under or on fallen logs, near  burrows, or on small animal trails. Traps were set for 3 days and 3 nights at each site and were baited asfollows: banana and peanut butter puree on snap traps,rolled oats and peanut butter mixture in Sherman and custom-made small mammal traps, banana slices, rolled oatandpeanutbuttermixtureinmedium-sizedTomahawk traps and this mixture mixed with sardines in the largeTomahawk traps. Traps were baited every morning and left open for 24 h. Each morning, traps were checked and species identified using Emmons (1990) and Reid (1997).Dead animals (from snap traps) were removed and liveanimals were either transported to the field station and collected as voucher specimens or else released   > 1 kmfrom the trapping area. Specimens were deposited and determinations completed at the museum in Davis usingHall (1981).Abundance was expressed as the number of indi-viduals/100 trap nights and, in some analyses, abundancewas broken down into the type of trap in which animalswere captured, i.e. small, medium or large, with snap,Sherman and custom-made all being treated as small. Thetotal number of mammal species inhabiting the Chiquibulthat might be captured in traps of these sizes ( n = 29) wastaken from Caro  et al  . (2001) and Kelly & Caro (in press)and supplemented with additional species caught duringthis study.Bird species richness and abundance were recorded at each site by walking a 1 km transect, 500 m in theforest interior and 500 m along forest edge (see Bibby,Burgess & Hill, 1992; Ralph  et al  ., 1993). Each transectwas surveyed on the 3 consecutive days that we trapped mammalsthroughdirectobservationandlisteningtocalls.Two observers (invariably A.E.) walked the transect for 2 h each day between 0700 and 1000 hours. Speciesidentification was aided by using Howell & Webb (1995).Abundance was expressed as the number of individualsseen per hour. The total number of species of birds inChiquibul ( n = 184 residents) was determined by addingthe number of species seen at all four sites augmented  by additional sightings on a road-edge transect, incidentalobservations around the field station and occasional mistnetting to verify species identification. We restricted thetotaltobirdspresentatthistimeofyearinordertoexcludeabsent migrants, the diversity and numbers of which canfluctuate from year to year and whose populations are notdependentupontheforestforreproductionandyearround resources.Scansorial, diurnal mammals were noted during bird transects and species and number of individuals wererecorded. Abundance was taken as the number of individuals seen per hour spent walking. Total Chiquibulspecies richness of scansorial diurnal mammals ( n = 9)was determined from Caro  et al.  (2001). RESULTSSpecies richness The proportion of frogs known to inhabit the Chiquibulwas low at all four sites, ranging from 33.3–19.0%(Table1).Therewasasignificantdifferenceinthenumber ofspeciesfoundatthefoursiteswiththetapirflagshipsitealong the river yielding more than eight times the averagenumber of frog species per unit effort. Two species werefound only at the non-flagship peccary site but this did notdiffer significantly from other areas.The proportion of Chiquibul phyllostomid bats found in the Maya Mountains was greatest at the peccary site(21.4%: Table 1). There was no significant difference inthe number of phyllostomid bat species captured per netnight across the four sites. The number of species uniqueto a site was greater at the peccary site than elsewhere.Theproportionofterrestrialmammalspeciesthatcould have been captured using our traps was relatively lowacrossallfoursites,approximately25%inallcasesexceptfor the tapir site which was slightly lower (Table 1). Therewere no significant differences in the number of speciescaptured at each site per unit effort. We failed to captureanymammaluniquetothejaguarflagshipsite;indeed,thissite had a significantly lower number of unique speciesthan did other sites.Theproportionofscansorialmammalsknowntoinhabitthe Chiquibul that was observed along transects ranged from 33.3% at the peccary site to 0% at the jaguar site (Table 1). The number of scansorial mammalsencountered per unit effort did not differ significantlyacross the four sites, nor did the number of species seenat only one site.Approximately one-third of the resident bird species inthe Chiquibul were observed at each site (Table 1) and thenumber of species seen per unit effort did not differ acrosssites. Turning to species that were found only at one site,there were significant differences across sites with morespecies being seen at the tapir site than at the other threecombined. Abundance In contrast to species richness, the number of individualsencounteredatthefoursitesdifferedsignificantlyformostgroups,yettherewasnoclearpatternacrosssites(Table2). Numbers of individual frogs differed significantly acrosssites and were far higher at the tapir site along the river.The number of individual phyllostomid bats also differed  between sites with most being captured at the peccary sitewhere nets were placed across a small creek.There was also a significant difference in the numbersof individual terrestrial mammals caught (Table 2). Whenthesewereseparatedintothosecapturedinsmall,medium
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks