Conservation Value of Remnant Riparian Forest Corridors of Varying Quality for Amazonian Birds ...

18 Pages·2008·0.64 MB·English

Contributed Paper
Conservation Value of Remnant Riparian Forest
Corridors of Varying Quality for Amazonian Birds
and Mammals
ALEXANDER C. LEES AND CARLOS A. PERES∗
CentreforEcologyEvolutionandConservation,SchoolofEnvironmentalSciences,UniversityofEastAnglia,NorwichNR47TJ,
UnitedKingdom
Abstract: Forestcorridorsareoftenconsideredthemaininstrumentwithwhichtooffsettheeffectsofhabitat
lossandfragmentation.Brazilianforestrylegislationrequiresthatallriparianzonesonprivatelandholdings
bemaintainedaspermanentreservesandsetsfixedminimumwidthsofriparianforestbufferstoberetained
alongsideriversandperennialstreams.Weinvestigatedtheeffectsofcorridorwidthanddegradationstatusof
37riparianforestsites(including24corridorsconnectedtolargesource-forestpatches,8unconnectedforest
corridors, and 5 control riparian zones embedded within continuous forest patches) on bird and mammal
species richness in a hyper-fragmented forest landscape surrounding Alta Floresta, Mato Grosso, Brazil. We
used point-count and track-sampling methodology, coupled with an intensive forest-quality assessment that
combinedsatelliteimageryandgroundtrutheddata.Vertebrateuseofcorridorswashighlyspecies-specific,
but broad trends emerged depending on species life histories and their sensitivity to disturbance. Narrow
and/or highly disturbed riparian corridors retained only a depauperate vertebrate assemblage that was
typical of deforested habitats, whereas wide, well-preserved corridors retained a nearly complete species
assemblage. Restriction of livestock movement along riparian buffers and their exclusion from key areas
alongsidedeforestedstreamswouldpermitcorridorregenerationandfacilitaterestorationofconnectivity
.
Keywords: Amazonia,habitatconnectivity,habitatfragmentation,habitatquality,riparianforest,tropicalforest,
wildlifecorridors
ElValordeConservacio´ndeCorredoresForestalesRipariosRemanentesconCalidadVariableparaAvesyMam´ıferos
Amazo´nicos
Resumen: Loscorredoresforestalesamenudosonconsideradoselprincipalinstrumentomedianteelcualse
atenu´anlosefectosdelap´erdidayfragmentacio´ndelha´bitat.Lalegislacio´nsilv´ıcolabrasilen˜arequiereque
todaslaszonasripariasenterrenosprivadosseanmantenidascomoreservaspermanentesydefineanchura
m´ınimadelosbosquesripariosamortiguadoresquedebenserretenidosalolargoder´ıosyarroyosperma-
nentes. Investigamos los efectos de la anchura y del estatus degradacio´n del corredor en 37 sitiosforestales
riparios(incluyendo24corredoresconectadosaparchesforestalesextensos,8corredoresforestalesnoconec-
tados y 5 zonas riparias control embebidas en parches de bosque continuos) sobre la riqueza de aves y de
mam´ıferosenunpaisajeforestalhiperfragmentadoenAltaFloresta,MatoGrosso,Brasil.Utilizamosm´etodos
de conteo por puntos y muestreo de huellas, adema´s de una evaluacio´n intensiva de la calidad del bosque
quecombino´ima´genesdesat´eliteydatosdeverificacio´nencampo.Elusodecorredoresporvertebradosfue
altamenteespec´ıfico,peroemergieronpatronesgeneralesdependiendodelashistoriasdevidadelasespecies
y de su sensibilidad a la perturbacio´n. Los corredores riparios angostos y/o muy perturbados retuvieron
unensambledevertebradosmuypobrequefuet´ıpicodeha´bitatsdeforestados,mientrasqueloscorredores
∗AddresscorrespondencetoC.A.Peres,[email protected]
PapersubmittedFebruary20,2007;revisedmanuscriptacceptedAugust27,2007.
1
ConservationBiology,Volume**,No.*,***–***
(cid:2)C2007SocietyforConservationBiology
DOI:10.1111/j.1523-1739.2007.00870.x
2 WildlifeUseofAmazonianCorridors
amplios,bienpreservadosretuvieronunensambledeespeciescasicompleto.Larestriccio´ndelmovimiento
deganadoalodeloscorredoresysuexclusio´ndea´reasclavealolargodearroyosdeforestadospermitir´ıa
laregeneracio´ndecorredoresyfacilitar´ıalarestauracio´ndelaconectividad.
Palabras Clave: Amazonia,bosqueripario,bosquetropical,calidaddeh´abitat,conectividaddeh´abitat,corre-
doresparavidaSilvestre,fragmentacio´ndeh´abitat
Introduction required to set aside a riparian forest strip along rivers
and perennial forest streams in the form of “permanent
The efficacy of wildlife corridors in facilitating animal protectionareas”(APPs).Theseriparianbuffersarepro-
movementsbetweenhabitatpatchesremainscontrover- tected by Brazilian federal legislation since 1965, which
sial (Rosenberg et al. 1997; Beier & Noss 1998; Bennett designatedfixedminimumwidthsofforestbuffersalong-
2003),butmostforesttaxaappeartorespondpositively sidewaterways(e.g.,30mforstreamsnarrowerthan10
totheirpresence.Corridorscanbeusedbyforestwildlife m [Co´digo Florestal 2001]), although levels of compli-
asmovementcorridors,conduitsthroughwhichanimals ance with minimum legal requirements are highly vari-
can disperse or commute between forest patches, and able(Resqueetal.2004).TheconservationroleofAPPs
habitatlinkages(foresthabitatthatsupportsresidentpop- presumablyincreasesinhighly deforestedregions,such
ulationsorlinkspopulationsamongpatches)(Rosenberg as the “Arc of Deforestation” of southern and eastern
et al. 1997; Lidicker 1999). Corridors should theoreti- Amazonia, which encompasses 524 municipal counties
cally facilitate gene flow between forest remnants and inhabitedby10,331,000peopleinthestatesofRondoˆnia,
reduce rates of stochastic extinction (Fahrig & Merriam MatoGrosso,Par´a,Tocantins,andMaranh˜ao.
1994) and the potential for deleterious genetic effects Intermsofwildlifehabitatrequirements,theminimum
brought about by inbreeding depression (Brown et al. width and structural preservation status of remnant for-
2004). estcorridorsformacontentiouspolicyareathatisyetto
Corridorshavebeendelimitedarbitrarilyinto2types: beinvestigated,butthesearelikelytobehighlyvariable
biodiversity conservation corridors (“biologically and acrossdifferenttaxadependingontheirrelativesensitiv-
strategically defined subregional space[s] selected as a ity to edge and area effects (e.g., Spackman & Hughes
unitforlarge-scaleconservationplanningandimplemen- 1995; Laurance & Laurance 1999). This is surprising be-
tation”)andbiologicalcorridors(“elongatedandcontin- cause corridors are often accessible and relatively easy
uouspatch[es]ofhabitatthatmaintain[s]connectivityal- tosample,andgroundtruthingdatacanbereadilyrelated
lowingthefluxofindividualsbetween2ormoreareas”) to spectral information derived from satellite imagery.
(Sanderson et al. 2003). Biodiversity conservation corri- Lima and Gascon (1999), who published the only study
dors obviously function as biological corridors but such on the utility of riparian corridors to Amazonian forest
“megacorridors” are financially and politically costly to wildlife, found no significant compositional differences
implement.Ontheotherhand,narrowforestcorridors, insmall-mammalandlitter-frogcommunitiesbetweenlin-
whichusuallycoursealongwaterways,areubiquitousin earremnantsandcontinuousforest.Theirresultssuggest
many tropical landscapes. These riparian corridors are thatcorridorsareimportantforatleastthesesmallverte-
either natural (e.g., gallery forests in tropical savannas) bratetaxathathavesmallarearequirements.
or anthropogenic features of the landscape (e.g., rem- We addressed the biodiversity value of tropical for-
nant riparian buffers set aside following deforestation), est corridors by investigating vertebrate-species occu-
yettheirroleinbiodiversityconservationremainspoorly pancyofremnantriparianbuffersinahyper-fragmented
understood. forest landscape of southern Amazonia. We compared
Current rates of tropical deforestation are unprece- patterns of species richness and composition between
dented, and this forest loss is most acute in Brazilian remnant riparian buffers and adjacent riparian sites em-
Amazonia, where by 2005 the total forest area cleared bedded in large areas of largely undisturbed continuous
had reached some 70 Mha (INPE 2006). Efforts to mit- forest. We focused on bird and mammal species that
igate forest conversion have focused on the creation use corridors of variable quality and traversing a matrix
of vast protected areas where there is the potential to of actively managed cattle pastures. Specifically we ex-
link them into reserve networks (Peres 2005; da Silva amined the minimum width and structural integrity of
et al. 2005). Nevertheless, forest retention in a grow- corridors required to maintain vertebrate assemblages
ing area of smallholdings and large private properties is compared to those of continuous primary forest and
also essential for the preservation of Amazonian biodi- whether the functional utility of corridors connected to
versity(Soares-Filhoetal.2006).Clearcuttingoperations largeforestpatchesishigherthanthatofentirelyisolated
by private landowners in Brazilian Amazonia are legally corridors.
ConservationBiology
Volume**,No.*,2007
Lees&Peres 3
Methods riparian forest sites (6 stations/site), including 24 con-
nected corridors, 8 unconnected corridors (isolated by
SamplingSites >300 m from the nearest forest patch), and 5 control
siteswithinlargepatches(11,030–144,700ha)ofundis-
Extensive road paving and several large agricultural re-
turbedprimaryforest.Corridorswerewidelydistributed
settlementprogramsduringthe1970scatalyzedmassive
throughout a ∼6000-km2 landscape and separated by
forest clearance in southern Amazonia. The countryside
>500 m (mean [SD] distance = 28.2 km [15.8] km; Fig.
around the town of Alta Floresta, Mato Grosso, Brazil
(09◦53(cid:4)S; 56◦28(cid:4)W; Fig. 1) is in the central Amazonian 1&SupplementaryMaterial).
Arc of Deforestation and is an ideal model landscape in
AvianandMammalSurveys
which to study the effects of habitat fragmentation and
perturbation (Peres & Michalski 2006). A complete de- Each site was surveyed twice, with a 75-d interval be-
scription of the study landscape is presented elsewhere tweensampling.Sixpoint-countsites(PCstations)were
(Michalski&Peres2005;Lees&Peres2006). located along each riparian corridor. The first was em-
FromMaytoOctober2005,weconducted444unlimi- bedded well within the source patch (>200 m from
ted-radius point counts at 222 sampling stations in 37 the forest edge), the second was 50 m from the patch-
corridor node, and the other 4 were located 200 m to
850 m apart (Fig. 1). We considered all species except
waterbirds (e.g., herons, rails), nocturnal species (e.g.,
owls, potoos, nightjars), and aerial insectivores (swifts
andhirundines).Wealsoassignedeachbirdspeciesto1
of4classesofforesthabitatspecificity(Stotzetal.1996)
(definedinTable1).Forafulldescriptionofthecatego-
rization scheme, a species list, and category scores see
LeesandPeres(2006).
Mammal presence and absence data were recorded
for diurnal primates on the basis of acoustic and visual
detection events and were obtained concurrently with
the avifaunal surveys during the PC sampling periods.
Presenceorabsenceoflargeterrestrialmammalspecies
(ungulates,carnivores,largerodents,andarmadillos)was
determined through intensive searches for tracks along
a 100-m riparian forest section just at the stream edge.
Boththeavianandmammalsurveysshouldberegardedas
conservativewithrespecttocorridor-widtheffects;how-
ever,surveysinnarrow(<100mwide)andunconnected
corridorswereexhaustive(seeSupplementaryMaterial).
Corridor,Patch,andLandscapeMetrics
Wemeasuredcorridorwidthperpendiculartothecorri-
dorlengthateachPCstationwithaHip-ChainandLand-
satimage.Wecombinedagroundtruthedassessmentof
forestqualitywithapixel-scaleremote-sensingapproach
todeterminethequalityofforestpatches.Followinga2-
stageunsupervisedclassificationoftheLandsatimage,we
Figure1. MapofstudyareaaroundAltaFloresta, wereabletounambiguouslyresolve8mutuallyexclusive
MatoGrosso,Brazil(09◦53(cid:4)S;56◦28(cid:4)W),showingthe land-cover classes ranging from closed-canopy forest to
sitessampledinconnected(solidcircles)and bareground(Michalski&Peres2005).Thenweextracted
unconnected(opentriangles)forestcorridorsand landscapevariablesfromtheLandsatimagewithFragstats
controlsites(solidsquares)embeddedwithin (version3.3,McGarigaletal.2002)andArcView(version
continuousforestsites.Forestandnonforestcoverare 3.2,EnvironmentalSystemsResearchInstitute,Redlands,
shadedgrayandwhite,respectively.Opencircles California).Wecalculatedthetotalareaofsourcepatches
denoteurbanareasand1isAltaFlorestaand2is connected to corridors by artificially eroding their nar-
Carlinda.Rectangularinsets(lowerpanel)show rowestconnections,usuallytocorridorbottlenecksnear
examplesofconnected(A)andunconnected(B) the patch node. Erosion of connections was always car-
corridorsandacontrolsiteincontinuousforest(C). riedoutacrossthenarrowestgroupsof15×15mpixels
ConservationBiology
Volume**,No.*,2007
4 WildlifeUseofAmazonianCorridors
Table1. Meanstructuralcharacteristicsandspeciesrichnessacrossthe37riparianforest(RF)sitesstudiedintheAltaFlorestaregion,including
24connectedcorridors(CC),8unconnectedcorridors(UC),and5riparianforestsitesembeddedwithinlargeareasofcontinuousforest(CF).a
Percentvariancec
Riparianforesttype(SD) ANOVAb explainedby
RF corridor
Variable CC UC CF F p type subset
Habitat
corridorwidth(log m)d 280.43(153.68)a 164.53(37.02)b — 9.04 <0.001 64.22 5.98
10
sourcepatchsize(log ha)d 3.37(0.80)b — 4.73(0.58)b — <0.001 90.76 9.24
10
spectralforestquality 7.98(0.76)a 6.52(0.76)b 8.43(0.22)b 8.67 <0.001 46.89 18.93
meanheightofverticald 260.63(31.75)a 214.12(49.82)b — 5.17 <0.001 28.53 26.03
profiles
treedensity(stems/ha) 256.52(83.45) 196.88(106.65) 293.0(86.19) 2.04 0.002 15.34 14.32
treebasalarea(m2/ha) 25.83(12.17)a 17.22(11.42)a 36.87(14.68)b 30.75 <0.001 26.46 19.07
nonpalmtreebasal 24.26(8.32)a 13.43(5.68)b 36.87(9.03)c 5.52 <0.001 38.82 14.72
area(m2/ha)
understorydensity 5.32(1.61)a 3.28(1.8)b 6.8(1.14)a 20.17 <0.001 22.84 12.47
canopycover(%) 71.38(24.43)a 38.22(34.60)b 83.72(11.93)a 6.83 <0.001 37.56 20.91
bambooabundance 0.60(0.83)a 1.06(1.29)b 0.73(1.00) 7.99 <0.001 0.0 50.98
Mauritiapalmabundance 0.69(0.92)a 1.81(0.77)b 0.20(0.21)a 9.67 <0.001 32.27 33.04
Speciesrichness
allbirds 100.70(21.19)a 70.62(12.88)b 141.4(6.38)c 6.40 <0.001 42.43 15.47
birds(S1)e 13.46(8.60)a 2.38(1.77)b 29.8(6.22)c 6.49 <0.001 47.34 12.13
birds(S2)e 39.36(13.29)a 17.87(4.64)b 65.8(1.92)c 10.10 <0.001 54.97 15.60
birds(S3)e 36.00(5.17)a 30.50(5.20)b 40.2(2.77)a 1.89 0.003 1.97 11.81
birds(S4)e 11.79(4.97)a 19.50(7.72)b 5.60(2.97)a 8.23 <0.001 48.93 16.18
allmammals 8.33(1.49)a 5.5(2.92)b 9.4(1.34)a 2.90 <0.001 21.28 10.61
largeterrestrialmammals 3.52(1.39)a 2.39(1.54)b 4.38(1.01)a 2.97 <0.001 24.37 16.40
primates 2.62(1.01)a 1.88(0.64)a 3.0(0.70)a 1.23 0.197 4.18 0.88
aGroupmeansofvariablesweretestedwithTukeymultiplecomparisons.Differentlettersrepresentsignificantdifferencesatα=0.05among
riparianforesttypeswheredifferencesamongsitesweresignificantwithinanalysisofvariance.
bThe analysis of variance (ANOVA) F tests are nested ANOVAs in which the 222 point-count (PC) stations were nested within each type of
remnantcorridororundisturbedcontrolsite.
cVariancecomponentanalysiswasusedtoestimatetheamountofvariabilitycontributedbyeachhierarchicalsitefactor(PCstationnested
withincorridorsubsets,whichwerenestedwithincorridortype).
dComparisonsbetweenonlyconnectedandunconnectedcorridors.
eHabitat-sensitivity classes for birds: S1, all strict forest understory and midstory species; S2, all remaining species dependent on primary
forest;S3,forestspeciesabletotoleratesecondaryorhighlydegradedforest;S4,primarilynonforestspeciesincludingscrubandopen-habitat
countrysidespecies.
inthemostdisturbedforest-coverclasssuchasyoungsec- terial). Estimates of forest height were derived by cali-
ond growth. A forest-quality index was then computed bratingthepixelheightswithmeasurementsofcorridor
for each PC station on the basis of the number of pix- heightwithaclinometer.Thisprovidedanotherestimate
elsrepresentingeachland-coverclass,incorporatingthe offorestperturbation:less-disturbedcorridorshadataller
nearest 10 pixels around each station. Structural forest andmoreuniformcanopyprofile,whereasdisturbedcor-
habitat variables quantified at each PC station included ridorswereoftenheavilyinvaginatedfollowingahistory
stand (palm and nonpalm) basal area, mean understory of selective logging and tree mortality induced by edge
density,andcanopycover.Inadditionwealsoquantified effects.
theabundanceofunderstorybamboo(Guaduasp.)and
Buriti palms (Mauritia flexuosa), intrusion by cattle,
DataAnalysis
andhuntingpressure(seeSupplementaryMaterial).
Wetook standardized digitalphotographs ofa100-m- Forestpatchsizeisakeypredictoroffaunaldiversityand
wide segment (centered at each PC station) of vertical explains 96% of bird (Lees & Peres 2006) and >90% of
corridorprofilesfrom120mperpendiculartotheforest mammal(F.Michalski&C.P.,unpublisheddata)species
edge. We used these images to determine corridor for- richness in the Alta Floresta forest fragments. Therefore
est quality. The images were then analyzed with Pixel we used the following strategy to examine patterns of
Counter(A.Etchells,UniversityofEastAnglia,Norwich, birdandmammalspeciesrichness.Firstweevaluateddif-
UnitedKingdom),whichcountedforest(dark)pixelsin ferencesbetweenall3riparianforesttypeswithone-way
one-pixel-widecolumnsacrosstheimageandcalculated and nested analysis of variance (ANOVA) and variance
a mean and SD for each image (see Supplementary Ma- componentanalysesinwhichmeasurementsateachPC
ConservationBiology
Volume**,No.*,2007
Lees&Peres 5
station (n = 222) were nested within each corridor (or significantlybetweenthetallerandmorestructurallyuni-
controlsite)associatedwithvariable-sizedpatches. formconnectedcorridorsandthelower-statureandmore
Second we modeled species richness (S) at the 192 degradedunconnectedcorridors(ANOVA;F=9.6,df=
PCstationsalongall32connectedandunconnectedcor- 31, p < 0.004). Across the 32 corridors, mean width
ridors with generalized linear mixed models (GLMMs) was positively correlated with spectral forest quality (r
in which S estimates were assumed to be nested within = 0.592, p < 0.001), corridor height (r = 0.425, p =
clusters (corridors) over which the random effects var- 0.015), nonpalm tree basal area (r = 0.317, p = 0.077),
ied. The GLMMs were fitted to model different forest, andcanopycover(r=0.473,p=0.006).Cattleintrusion
patch,andlandscapevariablesthatwereenteredasfixed occurred in 70% and 89% of all connected and uncon-
effects.Indifferentmodelsweincorporatedeithercorri- nectedcorridorplots,respectively,althoughwirefences
dor type or size of the source patch [log (x + 1)] as a were only erected in 16% and 2% of connected and un-
10
fixedeffect,butpatchsizeofunconnectedcorridorswas connected corridor plots, respectively. Cattle intrusion,
assumedtobezero. however, may have been suppressed or restricted in
Third we used analyses of covariance (ANCOVAs) to some plots by dense stands of bamboo (Guadua sp.),
test whether the slopes of species-corridor width rela- whichoccurredin24%ofconnectedcorridorplots,40%
tionships differed significantly between connected and ofunconnectedcorridorplots,and33%ofcontrolplots.
unconnectedcorridors.Fourthweevaluatedspeciesrich- Mauritia palms occurred in 42% of connected corridor
nessalongconnectedcorridorsatthe6PCstationsplaced plots, 90% of unconnected corridor plots, and 16% of
at varying distances from source patch nodes by stan- controlplots.
dardizing local species richness (S ) in relation to the
PCi
sourcepatchtowhichitwasconnected(S ).Changesin
SP
S((cid:2)S),expressedas(cid:2)S=(S /S )−1,therefore,ig-
PCi SP BirdAssemblages
noreddifferencesinthetotalnumberofspeciesretained
withinthe24sourcepatchesandcouldbemodeledwith We recorded 17,999 detections of 365 bird species dur-
a binomial error distribution across all connected sites ing 444 point counts. Mean corridor width was a signif-
with GLMs. Minimum GLMM and GLM models were fit- icantpredictorofbirdspeciesrichnesspercorridor(R2
ted within the R platform (Ihaka & Gentleman 1996) =0.393,p<0.001,n=32).Therewasacriticalwidth
andselectedbasedonasupervisedinformation-theoretic threshold of ∼400 m beyond which species accumula-
approach with the Akaike’s information criterion (AIC) tion did not increase significantly (Fig. 2). Other highly
(Burnham&Anderson2002). significantpredictorsincludedspectralforestquality(R2
Finally we investigated variation in species composi- =0.473,p<0.001)andthedistancefromthenearestof
tion among sites with nonmetric multidimensional scal- the2majorurbancenters(AltaFlorestaorCarlinda)(R2
ing ordinations (MDS; Clarke & Green 1988) with the = 0.372, p < 0.001). Less important but still significant
Bray–Curtis dissimilarity measure of presence–absence determinantsofbirdspeciesrichnessincludednonpalm
matricesandananalysisofsimilarity(ANOSIM;Faithetal. basalarea(R2=0.242,p=0.002),meancorridorheight
1987). We used the BIO-ENV procedure within PRIMER (R2 =0.111,p=0.035),andcanopycover(R2 =0.157,
(Carr1996)todeterminewhichcombinationofvariables p=0.014).
most influenced community composition (see Supple- Bird species were widely variable in their persis-
mentaryMaterial). tence in the 3 types of riparian forests. Some taxa (e.g.,
Red-bellied Macaw [Orthopsittaca manilata] and other
psittacids) were nearly ubiquitous across all sites and
Results
were more abundant in unconnected corridors because
ofthehigherabundanceofMauritiapalms,oneoftheir
CorridorForestStructure
keyfoodplants.Likewise,someriverinespecialistpasser-
Riparian forests (RF) around Alta Floresta were highly ines(e.g.,SilveredAntbird[Sclateria naevia])werefre-
heterogenous both in terms of their patch metrics and quently encountered in all 3 riparian forest types and
preservation status whether we considered unnested were only absent in the most degraded sites. Levels of
comparisons or nested ANOVAs in which the 222 PC species richness in control sites were far higher than
stations were nested within the 37 RF sites (Table 1). in either corridor types, and more species occurred in
Stand basal area was significantly different among the 3 connectedthaninunconnectedcorridors.Somespecies
typesofRFsites(one-wayANOVA;F=38.2,df=31,p conspicuouslyabsentfromunconnectedcorridorswere
<0.001).Remnantconnectedcorridorsretainedasignif- common in connected corridors (e.g., Black-tailed Tro-
icantlyhigherstructuralintegrityintermsoftheirwidth, gon [Trogon melanurus]), whereas others were com-
basal area, canopy structure, and height (estimated by mon in control sites and rare or absent in both cor-
verticalpixelcounts)comparedwiththosethathadlost ridor types (e.g., Cinereous Antshrike [Thamnomanes
connectivitytosourcepatches.Corridorheightdiffered caesius]).
ConservationBiology
Volume**,No.*,2007
6 WildlifeUseofAmazonianCorridors
Figure2. Relationshipsbetween
vertebratespeciesrichnessand
meancorridorwidthandforest
qualityforriparianforest
corridorsthatareeither
connected(shadedcircles)or
unconnected(opentriangles)to
largeforestpatchesandcontrol
siteswithincontinuousforest
patches(CF,dark-shaded
squares):(a,b)birdsand(c,d)
mammals.
Bird species richness was affected by different patch AccordingtotheBIO-ENVanalysis,themostimportant
and landscape characteristics in connected and uncon- grouping of variables predicting community structure
nected corridors and control sites, but responses were among all 37 sites were corridor width, spectral forest
highlyspecies-specific.Themostsignificantpositivepre- quality, Mauritia palm abundance, and cattle intrusion
dictorsofthenumberofprimaryforest-sensitivespecies (R=0.544).Excludingthe5controlsites,spectralforest
(classes 1–2) retained in riparian corridors were (in or- quality and bamboo and Mauritia abundance were the
derofimportance)corridorwidth,sizeofsourcepatch, most important combination of variables (R = 0.402).
and forest basal area (Table 2; Fig. 3), whereas Mauri- Unconnectedcorridorsandnarrowconnectedcorridors
tia palm abundance and cattle intrusion had a negative retained far fewer species than wide, connected corri-
effect.Converselytheless-sensitivespecies(classes3–4) dors and control riparian areas, and MDS scores indi-
werenegativelyaffectedbyforestcanopycover,butpos- cated they were more dissimilar from one another in
itivelyaffectedbyMauritiapalmabundanceandsource assemblage composition (Fig. 5). Similarly, community
patchsize.Forest-sensitivespeciesrespondedtobamboo compositiondifferedsignificantlyamongallripariansites
abundanceandcorridorheightandwidth,whereasless- (overallANOSIMR=0.501,p<0.05)andbetweenthe
sensitivespeciesweremorelikelytooccurinsitesoflow 3ripariantypes(overallANOSIMR=0.319,p<0.05).
forest quality, which contained a more heterogeneous
vertical profile. For the riparian sites within large forest
MammalAssemblages
patches, canopy cover was the only significant variable
retained for the most sensitive species, and there were We detected 794 tracks of 22 species of non primate
moreless-sensitivespeciesinlow-qualitypatches. mammals and 226 sightings of 5 primate species. Corri-
In connected corridors only, a higher fraction of the dorwidthwasasignificantpredictorofmammalspecies
speciesrichnessinadjacentsourcepatcheswaslostwith richness(allspeciescombined:R2=0.192,p<0.012,n
increasingdistancefromthesepatches(p=0.020),but =32;largeterrestrialmammals:R2 =0.147,p<0.017),
(cid:2)S was also significantly depressed at narrow corridor but not of primates alone (R2 = 0.076, p < 0.127). The
sites(p<0.001),whichcontainedlowercanopycover quality of the forest habitat was also a significant pre-
(p=0.002)andaspectralindexofpoorerquality(p= dictor of mammal species richness (R2 = 0.312, p =
0.042), particularly where cattle intrusion had regularly 0.001).Lessimportant,butstillsignificant,determinants
takenplace(p=0.016).Thisspeciesdecayalongcorri- included mean corridor height (R2 = 0.132, p = 0.023)
dorswasverypronouncedwithin50mofthepatchnode, andcanopycover(R2 =0.161,p=0.013).
butwasmoregradualwithincreasingdistancefromthe Aswithbirds,responsestocorridorswerehighlyspe-
sourcepatch(Fig.4). cies-specific (Table 2). Some species (e.g., small
ConservationBiology
Volume**,No.*,2007
Lees&Peres 7
Table2. Minimum,generalizedlinearmixedmodels(GLMMs)ofbirdandmammalspeciesrichnessin24connectedand8unconnectedcorridors,
accountingforpoint-count(PC)sitesnestedwithinclusters(corridors).a
Birds
primary- edgeand
Mammals
forest second-growth
all specialists tolerant all large primates
(358species) (207species) (151species) (18species) (13species) (5species)
Variable β p β p β p β p β p β p
Intercept −5.051 0.550−25.531 <0.001 22.828 <0.001 −1.634 0.248 −0.960 0.432 −0.262 0.563
Corridorwidth(m) 1.623 0.020 13.944 <0.001 −1.575 0.524 1.599 0.012 0.945 0.086
Patchsize(ha)b 0.113 0.059 1.497 0.019 0.251 0.519 0.188 0.048 0.203 0.030
Meanheightc 0.009 0.509 0.002 0.842 0.010 0.241 0.004 0.107 0.002 0.180 0.001 0.258
Spectralforestquality 0.546 0.352 −0.671 0.158 0.070 0.221
Treebasalarea(m/ha) 0.087 0.060 0.022 0.041 0.026 0.004
Understorydensity 0.037 0.104
Canopycover(%) 0.026 0.198 −0.036 0.021
Bambooabundance 1.099 0.092 0.641 0.207 0.277 0.014 0.330 <0.001
Mauritiapalms −1.129 0.109 −1.239 0.024
Cattleintrusion −1.460 0.415 −0.622 0.650 −0.597 0.063 −0.380 0.149 −0.216 0.169
Huntingscore 0.117 0.201
aCoefficients(β)andtheirrespectivepvaluesarelistedforallvariablesretainedinthebestmodels;blankcellsindicateexcludedvariables
(variablesnotincludedinthebestmodels).
bLog10transformed.
cMeanheight(inpixels)ofcorridorverticalprofilesbasedondigitalphotographs(seetext).
armadillos,Dasypusspp.)wereubiquitous,whereasoth- but rarer in unconnected corridors (e.g., paca [Agouti
ers(e.g.,Capybara[Hydrochoerus hydrochaeris])were paca]),whereasotherswereconspicuouslyabsentfrom
encounteredmorefrequentlyincorridorsthanincontrol bothcorridortypes(e.g.,spidermonkey[Atelessp.]).
sites.Nevertheless,encounterratesformostspecieswere Mammalspeciesrichnesswasaffectedbydifferentpre-
lower in corridors than in control sites. Some species dictorvariablesacrossthe3typesofsites.Inconnected
were common in control sites and connected corridors corridors, source patch size was the most important
Figure3. Relationshipsbetween
speciesrichnessandcorridor
widthfor4functionalgroupsof
birdspecieswithvaryingdegrees
ofhabitatsensitivity(sensitivity
classes;S1,allstrictforest
understoryandmidstoryspecies;
S2,allremainingspecies
dependentonprimaryforest;S3,
forestspeciesabletotolerate
secondaryorhighlydegraded
forest;S4,primarilynonforest
speciesincludingscruband
open-habitatcountrysidespecies).
Opentriangles,graycircles,and
blacksquaresindicate
unconnectedcorridors,connected
corridors,andcontrolriparian
sites,respectively.
ConservationBiology
Volume**,No.*,2007
8 WildlifeUseofAmazonianCorridors
Figure4. Changesinspecies
richness((cid:2)S)for(a)birdsand
(b)mammalsalongconnected
corridorsasafunctionof
distancefromtheirrespective
source-forestpatchnodes.
predictor, followed by corridor width, corridor height, sites(overallANOSIMR=0.115,p=0.28)butwassig-
canopy cover, bamboo abundance, and hunting pres- nificantly different among the 3 sampled riparian types
sure. In unconnected corridors, the only 2 variables re- (overallANOSIMR=0.509,p<0.001).
tainedintheGLMswereMauritiapalmabundanceand
SD of corridor height. Few variables had a strong effect
on the control sites, but those retained in the GLMs Discussion
included nonpalm basal area and understory density
(Table2). Our results show that many forest bird and mammal
FortheBIO-ENVanalyses,themostimportantgrouping species in southern Amazonia use riparian forest corri-
ofvariablespredictingmammalcommunitycomposition dors and that narrow remnant corridors fail to provide
acrossall37siteswerepatchsize,spectralforestquality, suitable habitat for many forest vertebrate species. Nar-
distancetothenearesturbancenter,bambooabundance, row,unconnectedcorridorstypicallyretainedonlyone-
andpresenceofcattle(R=0.368).Excludingthe5con- thirdofthebirdandone-quarterofthemammalspecies
trolsites,patchsize,distancetourbancenter,understory richness found in riparian forests within large forest
density, bamboo abundance, and spectral forest quality patches. Although corridor width was the most impor-
were the most important combination of variables (R = tantdeterminantofspeciesrichness,therewasastrong
0.343).TheMDSscoresshowedamorediffusescatterof interactionbetweenwidthanddegreeofforestperturba-
control sites, although the same broad trend of increas- tion,withwidercorridorsusuallyassociatedwithamore
ingsimilarityinassemblagecompositionoflargepatches intactcanopystructure.Yetnarrowripariancorridorsare
andcontrolsiteswasapparent(Fig.5).Communitycom- a predominant feature of many deforested landscapes
position did not differ significantly among all riparian in the humid tropics, including the expanding Arc of
Figure5. Vertebrateassemblage
compositionasafunctionof
forestcorridorwidthforpatch
sizefor(a)birds(stress=0.15)
and(b)mammals(stress=0.2).
Opentriangles,graycircles,and
blacksquaresindicate
unconnectedcorridors,connected
corridors,andcontrolriparian
sites,respectively.Circlesizeis
proportionaltoforestcorridor
widthandcontrolpatchsize,
whichwasthesignificant
predictorofthevariationin
multidimensionalscaling(MDS).
ConservationBiology
Volume**,No.*,2007
Lees&Peres 9
DeforestationofAmazonia(Resqueetal.2004).Thereis inginthenon-forestmatrixmayexplaintheiruseofun-
no evidence to suggest that differences in tree species connectedcorridors.Nevertheless,morearea-demanding
and faunal composition among sites were due to preex- speciessuchasthelarge-herdlivingwhite-lippedpecca-
isting differences in forest and soil types (see Peres & ries (Tayassu pecari), which require home ranges an
Michalski 2006; Michalski et al. 2007). An overwhelm- order of magnitude larger than those of collared pec-
ingproportionofchangesinspeciescompositionamong caries (Pecari tajacu) (Keuroghlian et al. 2004), were
sites is therefore assumed to result from differences in neverrecordedinisolatedcorridors.Carnivoresdiffered
patchandlandscapecharacteristics. significantly in their use of the 3 riparian forest types,
which may reflect differences in hunting pressure and
prey availability as much as matrix tolerance. Tayras
PatternsofCorridorOccupancy
(Eirabarbara)wereencounteredwithequalfrequency
Narrowcorridors(<200mwide)weremorevulnerable in all 3 riparian types. Small cats (Leopardus sp. and
toedgeeffectsthanwidercorridorandcontrolforestsites Pumayagouaroundi)wereencounteredatsimilarrates
andcontainednocoreforesthabitat.Thisrendersnarrow in connected corridors and controls but infrequently
corridorsmorevulnerabletoedgeeffects(Ferreira&Lau- in unconnected corridors, whereas signs of large cats
rance 1997; Cochrane & Laurance 2002), which can be (Puma concolor and Panthera onca) were also rare in
exacerbatedbytimberextraction,oftenleadingtostruc- unconnected corridors, uncommon in connected corri-
tural collapse of the corridor (Fig. 2). The persistence dors,andregularlyencounteredincontrolsites.Thetwo
within riparian corridors of some medium-to-high sensi- most frequently encountered primate species in uncon-
tivityriparianspecialists,suchasLong-billedWoodcreep- nectedcorridors—browncapuchins(Cebus apella)and
ers (Nasica longirostris), is encouraging. Nevertheless, duskytiti-monkeys(Callicebusmoloch)—werealsoleast
these species likely maintained narrow linear territories affectedbyfragmentationintheregionbecauseoftheir
along forest streams even within undisturbed areas and exceptional tolerance to habitat disturbance (Michalski
thereforecouldhavetheirarearequirementsmetinsuf- & Peres 2005). Hunting pressure did not significantly
ficiently wide (>200 m) and well-preserved corridors. affect large mammal species richness in corridors per-
Likewise,evenunconnectedcorridorsretainedsensitive haps because hunters in Alta Floresta could afford to be
riparianspecialistssuchasSilveredAntbirdsandtheen- highlyselectivebecauseofthehighavailabilityofbovine
demicGlossyAntshrike(Sakesphorusluctuosus). meat.
Conversely,otherspeciesandfunctionalgroupswere Castello´n and Sieving (2006) used radiotelemetry and
rarely recorded in any corridor type. This may be partly translocationstostudylandscapeusebyChucaoTapacu-
duetospecies-specificrequirementsforuplandforestbut los(Scelorchilus rubecula)andconcludedthatcorridor
isperhapsmorelikelyrelatedtoareaeffectsandedgein- protectionorrestorationandhabitatmanagementinthe
tolerance(Laurance&Bierregaard1997).Theabsenceof nonforestmatrixmaybeequallyfeasiblealternativesfor
Cinereous Antshrikes from many connected and all un- maintaining connectivity between forest patches. Nev-
connected corridors suggests that narrow corridors do ertheless, in the Alta Floresta region <30% of the avi-
not satisfy the area requirements for understory mixed- faunausedtheopen-habitatmatrix(S.Mahood&A.C.L.,
species flocks, although unaffiliated dispersing individu- unpublished data), which suggests that corridor protec-
als could theoretically move between patches through tionwherepossibleispreferabletomatrixmanagement.
connected corridors. Similarly, terrestrial insectivores Because our rapid surveys were biased against transient
such as the Black-faced Anthrush (Formicarius analis) birdsthatdonotholdaterritoryandthusareunlikelyto
and Ringed Antpipit (Corythopis torquatus) were un- vocalize, our results emphasize bird species capable of
commoninconnectedandabsentinunconnectedcorri- using corridors as part of their year-round home range,
dors,perhapsduetoterrestrialmammaloverabundance which in some cases included part of the source patch.
resulting in increased rates of nest predation for these This is consistent with the abrupt collapse in species
species(Stratford&Stouffer1999).Canopyflockswere, richnessatshortdistancesfromsourcepatches(Fig.4).
however, recorded much more frequently than under- Althoughcorridorsprovidefunctionalconnectivitybe-
story flocks in both connected and unconnected corri- tween patches, they may act as population sinks, with
dors. These flocks are more vagile and less sensitive to overspillfromsourcepatchesfollowedbypoorsurvivor-
fragmentation (Maldonado-Coelho & Marini 2004), so it ship within corridors (Henein & Meriam 1990; Crooks
isunsurprisingthattheyoccurredoveragreaterrangeof & Sanjyan 2006). For example, narrow forest-dividing
corridorwidths. corridors act as ecological traps (Gates & Gysel 1978)
Responsesbymammalspeciesweresimilarlyidiosyn- forforest-interiorNeotropicalmigrantsthatdonotavoid
cratic.Speciesoccurrenceinunconnectedcorridorsmay forest margins and experience higher levels of nest par-
be inextricably tied to matrix tolerance as much as area asitism and nest predation (Rich et al. 1994). Although
requirements. Observations of ungulates such as tapir narrow corridors may function as both sinks and traps,
andcollaredpeccariesregularlycrossingandoftenforag- theyarecertainlypreferabletonocorridors,considering
ConservationBiology
Volume**,No.*,2007
10 WildlifeUseofAmazonianCorridors
thelowtoleranceforopenhabitatsofmanyforestspecies widerthan10mifappropriatehabitatistobeprovided
andtheirpotentialuseoflinearforeststrips. forallbirdandmammalspeciessampled.
Unlikebirdsmostterrestrialmammalsdetectedduring Overcrowdingofspeciesandedgeeffectsarereduced
corridorsurveyswerelikelytransientindividualsbecause ascorridorwidthincreases,andwidercorridorsaccom-
the track surveys were not biased against transient indi- modate greater spatial heterogeneity. This provides a
viduals. For instance,we regularlysighted some species broader range of microhabitats that is often correlated
moving along the entire length of the sampled corridor with increased species richness (Lindenmayer & Nix
while the avian survey was being conducted (e.g., col- 1993;Bierregaardetal.2001).Capturingthishabitathet-
laredpeccaryherds).Corridorsitesalsoprovidedimpor- erogeneityiscriticalbecauseofautecologicaldifferences
tantfoodsources.Mauritiapalmsoccurredinalluncon- among species; for example, some bamboo or tall un-
nectedcorridorsandmostconnectedcorridors,thefruits floodedforesthabitatspecialistsarenotwelladeptatdis-
ofwhichareakeyfoodresourceforbothungulatesand persing through unfavorable habitats (Stratford & Stouf-
primates. fer1999).Ifsocietywishestomaintainbirdandmammal
species richness in fragmented forests in this area, we
recommendanurgentrevisionofthecurrentlyoutdated
Brazilian forest legislation, which should require the re-
PolicyImplications
tentionofwiderandless-disturbedforestcorridorsalong
Permanent protection areas (APPs) may be critical for watercourses. The persistence of riparian forests and
biodiversity conservation in Brazil, depending on the their associated faunal communities in deforested land-
landscape-scale density of the hydrographic network. scapeswill,however,requireacombinationofeffective
Forest remnants that buffer otherwise deforested ripar- enforcementofexistinglegislationviagroundpersonnel
ian areas are ubiquitous in the Alta Floresta landscape, and satellite monitoring systems, educational initiatives,
amountingtoameandensityof259mofriversandperen- andfinancialincentivestoprivatelandowners.
nialstreamspersquarekilometers.Atpresent,however,
there is considerable local variation between the legally
required minimum width, according to Brazilian legisla- Acknowledgments
tion (Law 7.803 of 18.7.1989), and the actual width of
forest buffers retained as APPs. The minimum width
This study was funded by a Natural Environment Re-
of30mforstreamsnarrowerthan10m(82%ofoursam-
searchCouncil(NERC)andasmallgrantfromtheCenter
pledcorridors)iswhollyinsufficientcomparedwiththe
for Applied Biodiversity Sciences at Conservation Inter-
critical-width threshold of ∼400 m our results indicate.
national.WethankV.daRivaCarvalhoandtheFundac¸˜ao
Buffers≥50mwidearelegallyrequiredforstreams10–
Ecolo´gicaCristalinoforcriticalsupportduringthestudy;
50mwide(Co´digoFlorestal2001).Asof2005,only14%
A.Etchellsforsoftwareprogramming;G.Arau´jo,K.Bar-
of a random set of 100 connected (mean width [SD] =
bieri, and F. Michalski for logistical help; S. Mayer for
260[320]m)andnoneof100unconnectedcorridors(90
providingbirdsoundrecordings;andallthelandowners
[55]m)thatwemeasuredthroughouttheAltaFlorestare-
andpeopleofAltaFlorestafortheirunreservedcoopera-
gion met this threshold value (Supplementary Material).
tion.
Hence,theusuallynarrowandheavilydegradedriparian
buffersremaininginourstudyregion,whicharetypical
ofotherdeforestedregionsofBrazil,areoflimitedusein
SupplementaryMaterial
termsofbiodiversityconservation.
Ripariancorridorsprovidemanyotherecosystemser-
Supplementary information on the methods (site selec-
vicestobothlandownersandwildlife.Manysmallholders
tion; avian and mammal surveys; corridor, patch, and
andcattleranchesacknowledgedthehydrologicalvalue
landscapemetrics;anddataanalysis)and3additionalfig-
offoreststripsadjacenttowatercoursesusedbylivestock
ures (photograph of the Alta Floresta landscape, sample
as drinking sites. Understory overbrowsing by cattle,
corridor profiles, and frequency distribution of corridor
however, had severe negative effects on terrestrial bird
widths) are available as part of the on-line article from
speciesbecauseitpreventedforestregeneration,which
http://www.blackwellsynergy.com/.Theauthorsarere-
is essential to restore structural and functional connec-
sponsibleforthecontentandfunctionalityofthesemate-
tivityofcorridors.Restrictinglivestockmovementalong
rials.Queries(otherthanabsenceofthematerial)should
riparianbufferswithfencesandexcludinglivestockfrom
bedirectedtothecorrespondingauthor.
keyareasalongsidedeforestedstreamswouldallowsec-
ondarysuccessionandfacilitateconnectivityrestoration
LiteratureCited
(Crooks & Sanjyan 2006). We recommend that riparian
strips should be >400 m wide (200 m on either side of
Beier,P.,andR.F.Noss.1998.Dohabitatcorridorsprovideconnectiv-
thestream)whereverpossible,particularlyalongstreams ity?ConservationBiology12:1241–1252.
ConservationBiology
Volume**,No.*,2007