DOC

Spatial

By Debra Cooper,2014-02-18 22:22
14 views 0
SpatialSpatia

    Spatial

    CurrentZoology56(6):670677,2010

    Spatialandtemporalvariationoffishassemblagesina

    subtropicalsmallstreamoftheHuangshanMountain

    YunzhiYAN,ShanHE.

    ,LingCHU,XiuyingXIANG,YanjuJIA,JuanTAO,

    YifengCHEN

    CollegeofLifeSciences,AnhuiNormalUniversity;ProvincialKeyLaboratoryofBioticEnvironmentandEcologicalSafetyin

    Anhui,Wuhu241000,Anhui,China

    CollegeofBiologicalScienceandEngineering,HebeiUniversityofEconomicsandBusiness,Sh~iazhuang050061,China

    InstituteofHydrobiology,ChineseAcademyofSciences,Wuhan430072,China AbstractSpatialandtemporalvariationoffishassemblageswereinvestigatedseasonallyfromMay2007toFebruary2008

    acrossl1studysitesinasubtropicalsmallstream,thePuxiStream,oftheHuangshanMountain.Alongthelongitudinalgradient

    fromheadwatertodownstream,fishspeciesrichnessandabundanceincreasedgradually,butthendecreasedsignificantlyatthe

    1owerreaches.ThehiP

    estspeciesrichnessandabundancewereobservedinAugustandthelowestinFebruary.Basedonanaly-

    sisofsimilarities(ANOSIM),fishassemblagesweresignificantlydifferentinspatialvariationbutnotintemporalvariation.A1

    thoughdifferenceswereobservedbothamongsitesandamongstreamorders,thelowerRvalueinordervariationsuggested

    streamorderwasnottheoptimalfactorexplainingthespatialvariationoffishassemblages.In

addition,damconstructiondidnot

    significantlyalterfishassemblagesinthesitesadjacenttoandimmediatelydownstreamtodams.Usingclusteranalysisand

    non

    metricMultiDimensionalScalinganalysis(NMS),assemblageswereseparatedintothreegroupsataBrayCurtissimilarity

    valueof42%:theupper,middleandlowergroups.Followinganalysisofsimilaritypercentagesofspeciescontributions(SIM

    PER),shiftsinoccurrenceorabundanceofcurriculus,zplatypus,R.bitterlingandA.fasciatuscontributedmosttothediffer

    encesamongstthethreegroups.StandardDeviationRedundancyAnalysis(RDA)suggestedthathabitatstructure(suchase/eva

    tion,substrate,andflowvelocity)contributedtothespatialandtemporalpa~emoffishassemblagesinthePuxiStream.Incon

    clusion.thefishassemblagesinPuxiStreampresentedsignificantspatialbutnottemporalvariation.Humandisturbancehasper

    hapsinducedthedecreaseinspeciesdiversityinthelowerreaches.However,nosignificantchangewasobservedforfishassem

    blagesinsitesfarfromandimmediatelydownstreamfromlow

    headdams[CurrentZoology56(6):670-677,2010].

    KeywordsAssemblagestructure,Lowheaddam,Streamfish,SpatialtemporalpaUem

    Maintainingspeciesdiversityofstreamdwelling

    fishesisoneoftheecologicalfunctionsofstreams.The

    identificationofpatternsofvariationinstream.dwelling

    fishassemblages,andtheirpotentialcausalmechanisms,

    isacentralthemeinstreamecology(Matthews,1998).

    Muchresearchhasshownthatfactorsinfluencingfish

    assemblagesinvolvethephysiochemica1environment.

    whichisspatiallyheterogeneousandtemporallyvari

    able.andbioticinteractionssuchascompetitionand

    predation(German,1988;HarveyandStewart,199l; Grossmaneta1.,1998;Dauwaltereta1.,2008).Along theheadwater-downstreamlongitudinalgradient,fish assemblagesoftenexperienceanincreaseinspecies richnessandabundance,mainlyresultingfromanin. creaseinhabitatcomplexityanddiversity,variationin therateoffishimmigration,andextinction(reviewedin Matthews,1986).However,anasymptoteoradecrease inspeciesrichnessisalsoobservedinthelowerreaches ofsomestreams,whichcouldbeduetogreaterpoilu tionlevels(e.g.,Oberdorffeta1.,1993).Smallstreams alsorepresentanecologicalgradientalongwhichup

    streamassemblagesarerelativelyvariablebutdown. streamassemblagesrelativelystable.Thisisassociated withthegeneralpa~emofupstreamenvironmentsbeing physicallyvariableandstructurallysimplewhiledown

    streamenvironmentsaretheopposite(Grossmaneta1., 1990).Theabovespatialtemporalpa~ernshavebeen

    testedalmostexclusivelyinmid.westemstreamsofthe UnitedStatesandMediterraneanstreamsinEurope.but theirgeneralityremainsunknown(Er6sandGrossman, 2005),especiallyinsubtropicalareaswhereannual changesintemperatureandprecipitationarequitediffer- entfromtemperateorMediterraneanareas.Recently, sometopicsregardingstreamfishassemblagesinsub

    ReceivedDec.16,2009;acceptedApr.16,2010 Correspondingauthor.E

    mail:yanyunzhi7677@126.com;yanyunzhi1976@yahoo.corn.cn 2010CurrentZoology

    YANYZeta1.:PuxiStreamfishassemblagevariation671

    tropicalareashavebeenreported,suchas:theeffectof selectivetimberextraction(Martin.Smith,1998)and seasonaltyphoons(Teweta1.,2002)onfishassemblages; theassessmentofhabitatqualitybyanindexofbiotic integritybasedonstreamfishassemblages(Bozzettiand Schulz,2004);andthetrophicstructureoffishalongan environmentalgradient(Kartharinaeta1.,2008).Todate, thespatio-temporalpattemofsubtropicalstreamfish assemblagesremainsrelativelyunknown.

    HuangshanMountainislocatedinthesouthernpart ofAnhuiProvince,China,andischaracterizedbya subtropicalmonsoonclimate.Atotalof122fishspecies, diversifiedintheirecologicalbehavior,havebeenob

    servedinthisarea(DiaoandShen,198l1.Impactsin. eludinghabitatdestruction,waterpollutionandover

    fishinghavecausedfishspeciesdiversitytoseverely declineinsomesmallstreamsofHuangshanMountain (Yaneta1.,2007).Here,weinvestigatedspatialand temporalvariationandcausalmechanismsoffishas. semblagesinthePuxiStream.oneofthefourmajor streamsatthenorthempartoftheHuangshanMountain, seasonallyoveraone.yearperiod.Ouraimswere:1,to determinespatiotemporalpattemsinfishassemblages inthePuxiStreamand21toidentifywhether1ow.head damsaffectfishassemblagesinsitesadiacenttoand immediatelydownstreamfromdams.

    1MaterialsandMethods

    1.1Studyarea

    ThesubtropicalmonsoonclimateofHuangshan Mountainischaracterizedbyasymmetricalseasonal

    temperaturesandprecipitation.Meantemperatures rangefrom22.1.CinJanuaryto27.5.CinJuly,with N

    anannua1meanOf17.8.C.Annualrainfallisplentiful (approximately2400mm/year),with79%ofrainfall duringspringandsummer(April-September)andless than5%duringthecolddrywinter.ThePuxiStream originatesfromtheLionApexandinflowsintothe TaipingLake,whichthenflowsintotheQingyiRiver,a tributaryofthe1owerreachoftheYangtzeRiver.Witha tota1lengthof30kmandacatchmentareaof158km. itisoneofthefourmajorstreamsinthenorthernpartof theHuangshanMountain.TheflowregimeinthePuxi Streamvariesseasonally,withalmostnoflowduring winterandhighflowinspringandsummer.emini- mumwatertemperature(closetooC,occursinJanuary, andthemaximum(approximately25.C)inJuly.Many 1owheaddams.1essthantwometers.werebuiltduring thelate20centuryforthepurposeofirrigationand domesticconsumption.Duetothelargenumberof residentsnearthemiddlereachofthisstream,thewater qualitydownstreamisworsethanupstream. 1.2Fishsampling

    Fishweresampledat11sitesinthePuxiStream(Fig. 1).Altitude,streamorder,adjacenttodamornot,and heightofthedamforthe1lsiteswereshowninT{le1. EachsitewassampledseasonallyduringMay,August, andNovemberof2007,andFebruaryof2008.Fish samplingwasstandardizedbystreamlengthandtime.A samplereachof100mwasfishedfor30minatevery

site,usingelectrofishing(CWB-2000P,China;12

    100HZ)bysinglepass,operatedbythesameperson withnettersthatfollowedazigzagrouteacrosseachsite. Fishwereidentifiedtospecieslevel(Rhinogobiussp. onlytogenusleve1),countedandretumedtothesite wheretheywerecaptured.

    7

    4,I\

    .

    /I,,,

    O-kn1,/'_B

    Fig.1SamplingsitesforourstudyonfishassemblagesinthePuxiStreamoftheHuangshanMo

    untain

    A.ThegeographicalpositionofthePuxiStreamatthenorthernofHuangshanMountain.?

    showingtheHuangshanMountain.B.Thesamplingsites onthePuxiStream,1-7showingthesitesnotadjacenttodams,I-IVshowingthesitesadjacentt

    odams.

    672Vl01.56No.6

    1.3Environmentalcharacterization

    Habitatateachsamplingsitewasdescribedusingten variables:elevation,wettedwidth,waterdepth,current velocity,discharge,watertemperature,dissolvedoxy? gen,pH,conductivity,andsubstratesize.Elevationwas measuredusingGPS.Streamwidthwasmeasuredalong fourtransects,regularlyspacedacrossthestreamchan- ne1.Waterdepth,dissolvedoxygen,pHandconductivity weremeasuredatfourequal-intervalpointsforeach transect.Currentvelocitywastakenat60%ofstream depthatthepointsofeachtransect.Alongeachtransect theproportionofsubstratecategories(particlesize1

0L-11TIITI,2:1--5mnl,3=55ITIITI,4=2550mill,5

    =

    50100n-fin,6=10O500mm,7=500L_l000mm,8=

    >1000mm)wasestimatedvisually,followingBaineta1. (1985).

    1.4Dataanalysis

    Analysisfocusedonquantifyingthespatialandtem- poralvariationsinspeciesrichnessandabundance,and identificationofenvironmentalvariablesexplaining variationsacrossthe1lstudysites.Sincethesampling effortwassimilaracrosssitesandseasons,species richnessandabundancewereuseddirectlyintheanaly- sis.Speciesthatoccurredatlessthantwositeswere excludedfromtheanalysistoavoidnegligibleweight

    ing.Speciesdominancewasassessedbytheimportance valueindex(IVI)asproposedbyKrebs(1989),using thefrequencyofoccurrenceandrelativeabundanceofa species.OnewayANOVAwasusedtotestthediffer-

    encesinspeciesrichnessandabundanceacrosssites duringfourseasonalsampling,andsignificantmain effectswerefurtheranalyzedusingFisher'sLSDwhen appropriate.SPSS5.0(SPSSInc.,Chicago,USA)was usedtoperformstatisticalanalysesandalphawassetat P<0.05.

    Discretetemporalandspatialpattemsinfishassem

    blageswereidentifiedusingPRIMER(v.5;Primer-E Ltd.2001).Richnessandabundancedatacollectedwere log(x+1)transformedtomeetassumptionsofmultivari

    atenormalityandtomoderatetheinfluenceofextremes inrichnessandabundance.Transformedsampledata

werethenusedtocreateaBrayCurtissimilaritymatrix

    calculatedforallpair-wisesamplecomparisons(Bray andCurtis.1957).Ananalysisofsimilarities(ANOSIM) wasusedtocomparefishassemblagesfromthe11sites acrossfourseasons.AtwowayAN0SIMwithoutrep-

    licationwasusedtotestforsiteandseasoneffects,since fishateachsiteweresampledonce,notrepeatedly,in everyseason.Atwo.waynestedAN0SIM(streamorder andseasonnestedwithdam,respectively)wasusedto testfordameffect,afterthefirstorderbeingdeleted fromanalysisfornodamwithinthisorder.Therela. tionshipsamongstassemblagesfromeachsiteare graphicallyrepresentedusingclusteranalysisanda non.metricMultiDimensiona1Scalinganalysis(NMS) (ClarkeandWarwick2001,.Thecontributionofeach speciestothedifferencesamongassemblagegroupswas identifiedusingSIMPERfClarkeandWarwick2001), ADetrendedCorrespondenceAnalysis(DCA)was performedtoidentifythestrongestgradientofassem- blagecompositionindependentoftheenvironmental variables.Sincethe1engthofgradientinDCAwas lowerthan4.standardDeviationRedundancyAnalysis (RDA1wasselectedfortheevaluationofthevariabil

    ityinassemblagestructureinrelationtothemeasured environmentalfactors.Themultivariatestatistical analysiswasperformedusingCANOCOversion4.5 (terBraakandSmilauer2002).Allsampledata(except pH)werelog(x+1)transformedtominimizetheeffect ofextremevalues.

    2Results

2.1Speciesdiversity

    Atotalof13,593fishwerecapturedinthePuxi Stream,representing24species(notincludingRhi

    nogobiussp.),10familiesand4orders;Cyprinidaewas themostnumerousfamilywith13species(Yanetal, unpublished).Speciesrichnesspersampleamountedto 7.74-3.5species.andabundance3034-244specimens. Palechub,Zaccoplatypus,wasthemostcommon(oc

    currenceinallsamples)andthemostdominantr2898% ofIVI)fish.Rhodeusbitterling,Rhinogobiussp., Acrossocheilusfascitus,Cobitissinensis,Abbottian rivularis,Misgurnusanguillicaudatus,Pseudogobio vaillanti,VanmaneniastenosomaandLiobagrusstyani werealsocommonr>40%ofsamples)anddominant YANYZeta1.:PuxiStreamfishassemblagevariation673 (>100%ofI?)species.Whereas,Crarus,Sarco

    cheilichysparvus,Opsarrichthysbidens,Ache~long- nathuschankaensis,Pseudobagrustruncates,Squalio. barbuscurriculus,SqualidusargentatusandMasta. cembelustierleatuswererestrictedr<40%ofsamples) butrelativelyimportantr>10%ofIVI)species.Another sevenfishwererarespeciesr<l0%ofIVI):Carassius auratus,Parbotiafasciata,Pseudorasboraparva,Mo

    nopterusalba,Ophicephalusargus,Silurusasotus,Sa. Kiangsiensis.

    Speciesrichnessgraduallyincreasedfromsite1-5, butdecreasedinsites6_7(Fig.2).Asignificantdiffer- encewasobservedinspeciesrichnessamongtheseven samplingsitesfarfromdams(OnewayANOVA,F6.21

22.37.P<0.05).Thespeciesrichnessofsite1wassig

    nificantlylowerthantheothersixsitesfP<0.05).In addition,thespeciesrichnessofsties4and5(notsig

    nificantlydifferentbetweenthetwosites,>O.05)were

    significantlyhigherthansites1,2,3and7(P<0.05). However,nosignificantdifferencewasobservedinspe

    ciesrichnessamongthefoursitesadjacenttodams (OnewayANOVA,F3.12=2.24,P>0.05).Fishabun

    dancealsovariedamongthe11sites(Fig.2).Overall, speciesrichnessandabundancewaslowestinFebruary (richness:5.27~2.79;abundance:124.82~142.15,and highestinAugust(richness:9.18~3.09;abundance: C

    .

    .J

    ?

    ?

    385.24~316.30).FishrichnessinNovemberr9.063.37)

    wasgreaterthaninMay(8.54~3.83),buttheopposite wasobservedinfishabundance(November,

    314.18~19.37;May,389.09~239.52).

    2.2Fishassemblages

    TwowayANOSIMwithoutreplicationtestingfor bothspatialandtemporalvariationsinfishassemblages suggestedthattherewasasignificantdifferenceinspa- tialvariation(R=0.68,P<O.011butnosignificantdif- ferenceinseasonalvariation=0.17.P>0.05). Two.waynestedANOSIMtestingfordamandstream ordereffectsshowedthatordersignificantlyinfluenced fishassemblages0.58,P<0.01),butnosignificant

Report this document

For any questions or suggestions please email
cust-service@docsford.com