生态

来源：易妖游戏网

PlantandSoil255:311–318,2003.

311

Seasonalchangesinthecontributionofrootrespirationtototalsoilrespirationinacool-temperatedeciduousforest

Mi-sunLee1,4,KaneyukiNakane2,TakayukiNakatsubo2&HiroshiKoizumi3

2Department

1DepartmentofEnvironmentStudies,GraduateSchoolofBiosphereSciences,HiroshimaUniversity,

Kagamiyama

1-7-1,Higashi-Hiroshima739-8521,Japan.ofEnvironmentalDynamicsandManagement,Gradu-ateSchoolofBiosphereSciences,HiroshimaUniversity,Japan.3RiverBasinResearchCenter,GifuUniversity,Yanagido1-1,Gifu501-1193,Japan.4Correspondingauthor∗

Received3May2002;acceptedinrevisedform21January2003

Keywords:carbon,cool-temperaturedeciduousforest,decompositionrate,rootrespiration,soilrespiration,soiltemperatureAbstract

Atrenchingmethodwasusedtodeterminethecontributionofrootrespirationtosoilrespiration.Soilrespirationratesinatrenchedplot(Rtrench)andinacontrolplot(Rcontrol)weremeasuredfromMay2000toSeptember2001byusinganopen-ﬂowgasexchangesystemwithaninfraredgasanalyser.Thedecompositionrateofdeadroots(RD)wasestimatedbyusingaroot-bagmethodtocorrectthesoilrespirationmeasuredfromthetrenchedplotsfortheadditionaldecayingrootbiomass.ThesoilrespirationratesinthecontrolplotincreasedfromMay(240–320mgCO2m−2h−1)toAugust(840–1150mgCO2m−2h−1)andthendecreasedduringautumn(200–650mgCO2m−2h−1).Thesoilrespirationratesinthetrenchedplotshowedasimilarpatternofseasonalchange,buttherateswerelowerthaninthecontrolplotexceptduringthe2monthsfollowingthetrenching.Rootrespirationrate(Rr)andheterotrophicrespirationrate(Rh)wereestimatedfromRcontrol,Rtrench,andRD.WeestimatedthatthecontributionofRrtototalsoilrespirationinthegrowingseasonrangedfrom27to71%.TherewasasigniﬁcantrelationshipbetweenRhandsoiltemperature,whereasRrhadnosigniﬁcantcorrelationwithsoiltemperature.Theresultssuggestthatthefactorscontrollingtheseasonalchangeofrespirationdifferbetweenthetwocomponentsofsoilrespiration,RrandRh.

Abbreviations:RD–carbonemissionduetothedecompositionofresidualroots;Rh–heterotrophicrespirationrate;Rr–rootrespirationrate;NEP–netecosystemproduction;NPP–netprimaryproductivityIntroduction

StudieshavesuggestedthatforestscanbeeithersourcesoforsinksforatmosphericCO2(Malhietal.,1999;Saxeetal.,2001;Valentinietal.,2000).Carbonnetbalance(NEP–netecosystemproductiv-ity)inforestecosystemsisdeterminedasthebalancebetweennetprimaryproduction(NPP)ofvegetation

∗Presentaddress:CenterforGlobalEnvironmentalResearch,

andheterotrophicrespiration(Rh)ofsoil:

NEP=NPP–Rh

(1)

Respirationduetosoilheterotrophs–fauna,bacteriaandfungi–(Rh)isthedifferencebetweentotalsoilrespiration(Rs)andtheactivityofautotrophicrootsandassociatedrhizosphereorganisms(Rr)(EdwardsandHarris,1977):

Rh=Rs–Rr

(2)(3)

NationalInstituteforEnvironmentalStudies,16-2Onogawa,Tsukuba,Ibaraki305-8506,Japan.FAXNo:+81-29-850-2960.e-mail:lee.misun@nies.go.jp

NEP=NPP–(Rs–Rr)

Preciseassessmentofthesecomponentsisimport-antforquantifyingNEP.Althoughtotalandhetero-

312

trophicrespirationhavereceivedconsiderableatten-tioninrecentdecades,muchlessisknownaboutthecontributionofrootrespirationtototalsoilres-piration.Thecontributionofeachcomponentneedstobeunderstoodinorderthattheimplicationsofenvironmentalchangeforsoilcarboncyclingandsequestrationcanbeevaluated(Hansonetal.,2000).Thecontributionofrootrespirationtototalsoilrespirationisdifﬁculttodetermine,asreﬂectedbythewiderangeofpublishedestimatesforforestsoils(10–90%:Bowdenetal.,1993;Coleman,1973;Ed-wards,1991;Nakaneetal.,1996;Hansonetal.,2000;Ohashietal.,2000;ThierronandLaudelout,1996).Althoughsomeofthisvariabilityreﬂectsdifferencesamongtypesofecosystems,aconsiderableproportionofitprobablyoriginatesfromthevarietyofmeas-urementtechniquesused,eachwithauniquesetoflimitations(Rochetteetal.,1999).Methodsforsep-aratingrootrespirationfromtotalsoilrespirationhavebeenreviewedbySinghandGupta(1977)andHansonetal.(2000)indetail.

Thetrenchingmethodofmeasuringrootrespira-tionisrelativelysimple,anditsuseiscommoninforestecosystems(Hansonetal.,2000;Rochetteetal.,1999).Inthismethod,therootsexistinginagivenareaareseveredattheplotboundarybutnotremoved.Therootrespirationrateisestimatedfromthediffer-enceinsoilrespirationratesbetweenthetrenchedplotandacontrolplot.However,oneofthebiggestcon-cernsistheinﬂuenceofresidualdecomposingrootsleftinthetrenchedplotsandtheircontributiontosoilrespiration.

Anadditionalchallengetoestimationisthepoten-tialseasonalityofrootactivity,whichmaycausethecontributionofrootrespirationtototalsoilrespirationtochangeseasonally.Studieshaveshowndifferencesbetweenthegrowingseasonandthedormantseasoninthecontributionofrootrespirationtototalsoilrespir-ation(Edwards,1991;RochetteandFlanagan,1997).However,littleinformationisavailableonthecon-tributionofrootrespirationtototalsoilrespirationduringthegrowingseason.

Thisstudyaimedtorevealseasonalchangesinthecontributionofrootrespirationtototalsoilrespirationinacool-temperate,deciduous,broad-leavedforestincentralJapan.SoilrespirationratesintrenchedandcontrolplotsweremeasuredfromApriltoNovem-ber2000andfromMaytoSeptember2001byusinganopendynamicchambertechniquewithaninfraredgasanalyser(IRGA).Carbonemissionduetothede-compositionoftheresidualrootswasestimatedbya

root-bagmethod.Onthebasisofthesedata,weex-aminedseasonalchangesinestimatedrootrespirationandheterotrophicrespiration.MaterialsandmethodsStudysite

Thestudysiteliesonthenorth-eastslopeofMtNorikura,GifuPrefecture,centralJapan(36◦80󰀁N,137◦26󰀁E,1430masl).Theareaisclassiﬁedascool-temperateandislocatedinatransitionalzonebetweenthePaciﬁcsideandtheJapanSeasideofJapan.From1980to2000,theannualmeanairtemperaturewas6.1◦Candtheannualmeanprecipitationwas2175mm(TakayamaExperimentalStation,GifuUniver-sity;1342masl).ThesiteiscoveredwithsnowfromDecembertoApril.TreesproduceleafbudsatthebeginningofJune,andtheleavesbegintosenesceinearlyOctober.

A40-year-oldsecondary,deciduous,broad-leavedforestcomposedmainlyofQuercuscrispulaBlumeandBetulaermaniiCham.growsatthesite.Inareaswithbrownforestsoils(DystricCambisols,FAO,1990),theforestﬂoorvegetationconsistedofSasasenanensisRehd.,abamboo.Thestudysitewasestab-lishedin1993forthelong-termmeasurementofCO2ﬂuxinAsiawiththecooperationoftheNationalInsti-tuteofAdvancedIndustrialScienceandTechnologyandGifuUniversity.

Thestudysiteisbeingusedforcontinuous,tower-based,eddyﬂuxcorrelationmeasurements(e.g.,Sai-gusaetal.,2002;Yamamotoetal.,1999).Ground-basedecologicalmeasurements(e.g.,Kawamuraetal.,2001;Marikoetal.,2000)wereundertakentocomplementthetower-basedmeasurementsofCO2exchange.Trenching

Thetrenchplotwasestablishedadjacenttothecontrolplot(each5×5m)atthestudysiteinmidMay2000.Foursubplots(each60×60cm)wereestablishedinthetrenchplot(Figure1).Thesubplotswerepreparedbymakingverticalcutsalongtheboundariesto40cmbelowthegroundsurface(approximatelythebottomoftherootzone)withasteelknife,severingallroots.Therootswerenotremoved.Piecesof0.5-cmthickpolyethyleneboardwereinsertedintotheverticalcutstoinhibitrootregrowth,andtheirjoinsweresealedwithextrastripsofboard.

313

Figure1.Layoutofmeasurementplots(left)andtrenchingmethod(right).Eachplotcontainsfourpoints.

Tokilltherootsintheplots,theabovegroundpartsofallplantsgrowingintheplotswerecutoffatthelittersurface.Newseedlingsandregrowthfromtherootswereperiodicallyclippedwhennecessary.

ApreliminarystudyindicatedthatsubterraneanrootsofSasacouldsurviveandmaintaintheirres-piratoryactivityafterthetrenching.Therefore,weapplied1–2mLofa10%solutionoftheherbi-cidedicamba(dimethylaminesaltof2-methoxy-3,6-dichlorobenzoicacid;HodogayaKagakuKogyoCoLtd,Tokyo,Japan)toallcutstemswithasyringeonceaweekinMay2000andonceamonthinJuneandJuly2000.Mostofthesubterraneanstemsbegantodecomposeby2monthsaftertheapplicationoftheherbicide.

Soilrespirationmeasurements

SoilrespirationwasmeasuredonceortwiceamonthfromApriltoNovember2000andfromMaytoSeptember2001.InMay2000,whenthetrenchingwascarriedout,measurementsweremadeeveryweek.Thesoilrespirationratewasdeterminedusinganopendynamicchambertechniquesimilartothatde-scribedbyMarikoetal.(2000)andLeeetal.(2002).Pairedmeasurementsweremadesimultaneouslybythetwooperatorsfromtwomeasurementsubsystems,onepertreatment.Eachsubsystemcomprisedaref-erencegaslineandfouridenticalsamplegaslines(oneperchamber),anIRGAinabsolutemode(Li-Cor6252,Lincoln,NE,USA),adataloggerandapersonalcomputer.Ambientair(50cmfromthesoilsurface)waspassedthroughthechambersatarateof0.5L

min−1.Carbondioxideconcentrationsoftheairbeingpumpedinto(seebelow)andwithdrawnfromeachchamberinturn(n=4)weremeasuredbyIRGA.Onemeasurementcycleof25minwasrepeatedfor24–48h.MethodsformeasurementsystemofsoilrespirationhavebeendetailedbyLeeetal.(2002).

Eachchamberconsistedoftwoparts.Thelowerbody,withtwoportsfortheinletandoutletofair,wasa15-cmhighPVCcylinderwitha21-cminternaldiameter.Thecross-sectionalareaofthechamberwas346cm2.Thebottomedgeofthecylinderwaspushed4cmintothesurfacelitterlayer.Thisdepthwasad-optedtopreventgasleakageandtominimiserootdissection.Liveorstandingdeadvegetationenclosedwiththecylinderwascarefullyavoided.ThelowerbodywasleftatthestudysitefromApriltoNovembersoastominimisesoildisturbance.Theupperpart,a2.5-cmhighPVClid,wasplacedonthetopofthebodyimmediatelybeforemeasurementswerestarted.Soiltemperaturewasmeasuredineachchamberduringtheﬂuxmeasurement(5–10cmsoildepth).Continuousmeasurementsofsoiltemperatureweretakenattwopointsneartheﬂuxmeasuringpointsat30-minintervals(10cmsoildepth)overtheentirestudyperiod.

Soilwatercontentswasmeasuredincontrolandtrenchedplotsatasoildepthof12cmbelowthetopofthelitterlayerbyusingtimedomainreﬂecto-metry(TRIME-EZ;TohokuElectronicIndustrialCoLtd,Sendai,Japan).

314

Rootbiomass(Br)andrelativelossrateconstant(k)Toestimaterootbiomass(Br),threeplots(each1×2×0.5m)wereestablished40–50mfromthecon-trolplot.TherootswereexcavatedinOctober1999(n=2)andinMay2000(n=1)anddividedintotreerootsandSasaroots.Thetreerootswereclassiﬁedintothreesizeclasses:ﬁne(<2mmdiameter),medium(2–10mm)andthick(>10mm).TheSasarootswereclassiﬁedintosubterraneanstemorﬁneroots.Sub-samplesofrootsweredriedat85◦Cfor2weekstoestimatethedrymassofthetotalrootsample.

RootsamplescollectedinMay2000wereplacedinbagsmadeof1-mmnylonmeshtoestimatethere-lativelossrateconstant(k)inthesoil.Thebagsusedforthethicktreerootswere50×50cmandthebagsusedfortheotherrootsizeclasseswere25×25cm.Atotalof58bagswereprepared(15ﬁne,9medium,10thick,9subterraneanstem,and15ﬁneSasaroots).Theywereburiedatasoildepthof10–20cminMay2000,andthedecreaseindrymassoftherootswasmeasuredat83days(5,3,4,3and5,respectively)and181days(10,6,6,6and10,respectively)afterburial.

MasslossfromdecomposingrootswasanalysedbyOlson’s(1963)standardexponentialdecayfunctionX/X0=ae−kt,whereX/X0isthefractionofinitialmassremaining(X=rootmassattimet,X0=initialmass),tistime(year−1),andkistherelativelossrateconstant(theslopeofthelinearregressionﬁtforrootsofeachclass;a=intercept).

Estimationofrootrespirationrates(Rr)

Therootrespirationrate(Rr)wascalculatedbythefollowingequation:

Rr=Rcontrol–(Rtrench–RD),

(4)

whereRcontrolisthesoilrespirationrateinthecontrolplot,Rtrenchisthesoilrespirationrateinthetrenchedplot,andRDisthecarbonemissionrateduetothedecompositionofresidualrootsinthetrenchedplot.Deadrootsaredecomposedbysoilmicrobesandanimalsandgraduallyturnintohumus.Theorganicmatterderivedfromdeadrootsmaynotbeabletomovedownwardsuntiltherootlitterhasbeenhumi-ﬁedtoaconsiderableextent(Nakane,1978).AlthoughRDhasadirectrelationshipwiththerelativelossrateconstant(k),thelattervalueincludesnotonlydecompositionbutalsothehumus-formingprocess(transportation).Weused2/3asthedecomposition(mineralisation)rate,asproposedbyNakane(1978)

orNakaneetal.(1996)inthecalculation.Rootde-compositionrates(ν)werecalculatedbythefollowingequation:

ν=0.k

(5)

Carbonemissionduetothedecomposition2ofresidualrootsofeachsizeclass(Rd)(gCm−day−1)atagiventimetwascalculatedasfollows:

Rd=Br(ae−ν(t−1)–ae−νt)

(6)

RDisthesumoftheRdvaluesofeachsizeclass(=󰀁RDRd).

Heterotrophicrespiration(Rh)wascalculatedasfollows:

Rh=Rcontrol–Rr(7)

Statisticalanalysis

Theeffectsoftreatment(controlversustrench)andmeasuringtimeonthesoilrespirationratewereas-sessedbytwo-wayANOVA.Comparisonofregres-sionlinesbetweenrespirationrateandsoiltemper-aturewasdonebyANCOVA.SigniﬁcanceforallstatisticalanalyseswasacceptedatP=0.05.AllstatisticalanalyseswereperformedwithStatView5.0(SASInstituteInc.Cary,NC,USA).Results

Seasonalchangesinsoilrespirationrates

ThesoilrespirationratesinthecontrolplotincreasedfromMaytoAugustandthendecreasedduringau-tumn(Figure2).Themaximumratesrecordedin2000and2001were850and1200mgCO2m−2h−1,respectively.Thesoilrespirationratestendedtobehigherin2001thanin2000,possiblybecauseofslightlyhighersoiltemperaturesin2001(Figure2).Beforethetrenching,therewasnosigniﬁcantdif-ference(P=0.187)inthesoilrespirationratebetweenthecontrolandtrenchedplots(Figure2).Soilrespirationratesinthesameplotsdeterminedinthepreviousseason(fromSeptembertoDecember1999)didnotdiffersigniﬁcantly(resultsnotshown).Thetrenchingcausedasigniﬁcantincrease(P<0.001)inthesoilrespirationratethatlastedfor1–2months.Afterthat,thepatternofseasonalchangeinsoilres-pirationratesinthetrenchedplotwassimilartothatinthecontrolplot,buttherateswere16–37%lowerthaninthecontrolplot(Figure2).

315

Figure2.Seasonalchangesinsoilrespirationrateinthecontrolandtrenchedplots(mean±SD).Thearrowindicatestrenchingtime.Anasteriskdenotesasigniﬁcantdifference(P<0.001).Table1.Rootbiomass(gCm−2)ineachrootsizeclass

Rootsize

Tree

Thick(>10mm)Medium(2–10mm)Fine(≤2mm)Subtotal

SubterraneanstemFineSubtotal

Rootbiomassa(gCm−2)204±71136±4699±25

439

100±25134±13

234673

Estimationofrootrespirationrates(Rr)

Intheﬁrstyearoftrenching(2000),thecontributionofRDtototalsoilrespirationwas14–52%(mean16%),althoughthevaluewashighestinNovember(52%;Figure3).Inthesecondyearoftrenching(2001),thecontributionofRDtoRswas7–8%,exceptinMay(26%).

TheestimatedRrdecreasedfromJulytoNovemberin2000(Figure3).Highrespirationratesaftertrench-ingdidnotallowustoestimateRrbeforeJuly2000.ThepatternofseasonalchangeinRhwassimilartothatoftotalsoilrespiration(Figure2),withaﬁrstpeakinAugust(3.61gCm−2day−1)andasecondpeakinOctober(2.gCm−2day−1).Inthesecondyearoftrenching(2001),thevalueofRrpeakedinearlysum-mer(3.03gCm−2day−1)andrapidlydecreasedfromJulytoSeptember.Incontrast,nomarkedreductioninRhwasobservedthroughoutsummer2001(Figure3).Therewasasigniﬁcantexponentialrelationship(r=0.83,P<0.05,n=10)betweenRhandsoiltemperature,whereasRrhadnosigniﬁcantcorrelation(r=0.37,P>0.05,n=10)withsoiltemperature(Figure4).

ThecontributionofRrtototalsoilrespirationin2000rangedfrom32–48%,exceptinNovember,whenitwas71%.Thecontributionin2001wassimilar,27–39%,althoughanexceptionallyhighvalue(63%)wasobtainedforMay.ThecontributionofRrtototal

Sasa

Totalrootbiomass

aValuesaremeans±SE(n=3).

Rootbiomass(Br)andrelativelossrateconstant(k)Thetotalrootbiomassintheexcavatedplotwases-timatedtobe673gCm−2,about35%ofwhichwasduetoSasaroots(Table1).FinerootsoftreesandSasaconstitutedabout15and20%ofthetotalrootbiomass,respectively.

Morethan90%oftherootbiomasswasintheupper20cmofthesoil.Atanearbysite,morethan99%oftheﬁnerootbiomasswasintheupper40cm(HashimotoandHyakumachi,1998).

Thevaluesoftherelativelossrateconstant(k)andthecorrelationcoefﬁcient(r2)calculatedfromahalf-year(181days)lossofrootsineachrootsizeclassareshowninTable2.

316

Table2.Exponentialrelativelossrateconstant(k)andthecoefﬁcientofdetermination(r2)calculatedfromahalf-year(181days)lossofrootsineachrootsizeclass.

Relativelossrate

constantk(year−1)

0.710.830.760.590.79

Rootsize

Intercept

a0.950.980.940.980.99

Correlationcoefﬁcient

0.860.990.830.970.99

Tree

Sasa

Thick(>10mm)Medium(2–10mm)Fine(≤2mm)SubterraneanstemFine

Figure3.Seasonalchangesinheterotrophicrespirationrates(Rh),rootrespirationrates(Rr)androotdecompositionrates(RD).

soilrespirationdeterminedinJuly2000(70daysaftertrenching)wasalmostthesameasinJuly2001(39%).

Discussion

Thebasicassumptionofthetrenchingmethodisthatrespiratoryactivityofrootsinthetrenchedareaiscompletelysuppressedbecauseofthelackofanen-ergysupply.However,severalstudieshaveshownthatrootsexcisedfromthemainstemcansurviveandmaintainrespirationforsometimeafterexcision(Tateetal.,1993;Uchidaetal.,1998).Moreover,inmanycases,signiﬁcantincreasesinroot(orsoil)respirationrateswereobservedimmediatelyafterrootexcision(ortrenching),whichcanbeattributedtoin-jurybyexcisionanddisturbancetosoil(Eweletal.,1987;Uchidaetal.,1998).Inthisstudy,similarhighsoilrespirationratesaftertrenchingwereobserved;

Figure4.Relationshipsbetweenroot(Rr)andheterotrophic(Rh)respirationandsoiltemperature.Thelinewasﬁttedwithanexpo-nentialfunctionforRh.

soilrespirationratesofthetrenchedplotwerehigherthanthoseofthecontrolplotfor1–2monthsaftertrenching.

Afterthatperiod,thesoilrespirationrateinthetrenchedplotwassigniﬁcantlylowerthaninthecon-trolplot,suggestingthattrenchingreducedrootres-piratoryactivity.Althoughwecouldnotexaminetheviabilityofrootsinthetrenchedplots,weconductedapreliminaryexperimentontheeffectsoftrenchingandherbicideonrootviabilityinthesameforeststand.Therootsturnedblackandbegantodecompose2–3monthsaftertrenching.Therefore,weassumethatrootrespirationwasnegligibleinthetrenchedplotby3monthsaftertrenching.Thisassumptionwassupportedbythefactthatthecontributionofrootres-pirationtototalrespirationdeterminedinJuly2000(70daysaftertrenching)wasalmostthesameasthecontributioninJuly2001.

Newrootgrowthintothetrenchedplotfrombelowthebaseofthepartitionswasnotpreventedmechanic-ally.However,since99%oftherootbiomasswasintheupper40cmofthesoil,theeffectofrootgrowthinthedeepsoillayerwasassumedtobesmall.

Onepossibleimpactoftrenchingonsoilphysicalparametersisthattrenchingincreasessoilmoisturebyreducingtranspiration.Therefore,weexaminedsoilwatercontentsincontrolandtrenchedplotsatasoildepthof12cmbelowthetopofthelitterlayer.Therewasnosigniﬁcantdifference(P=0.214)insoilwatercontentbetweentheplots.

Theorganicmatterderivedfromdeadrootsisgraduallydecomposed.Thismayincreasethesoilrespirationrateinthetrenchedplot.However,thein-ﬂuenceofdecomposingrootsonsoilrespirationhasbeenrarelyexaminedinprevioustrenchingstudies(Bowdenetal.,1993;Eweletal.,1987;Faheyetal.,1988).Ourresultsindicatethatthecontributionofcarbonemissionthroughthedecompositionofrootswasabout16%inthegrowingseasonoftheﬁrstyear(July–October2000),butdroppedtoabout7%inthesecondyear(2001).ThisagreeswellwiththedataofOhashietal.(2000),whoestimatedthatRDcontrib-utedabout8%ofsoilrespirationinthesecondyearoftrenchinginaJapanesecedar(CryptomeriajaponicaD.Don)plantation.

Wemayhaveoverestimatedrootrespirationandthusunderestimatedheterotrophicrespiration,becauseweneglectedtheturnoverofﬁneroots.Thedeathofﬁnerootsmaystimulatemicrobialactivity,therebyincreasingheterotrophicrespiration.Nakane(1978)indicatedthatthedecompositionrateofﬁnerootswas

317

equivalenttoabout20%oftheirbiomassinacool-temperateforestinJapan.Atrialcalculationofcarbonemissionduetorootturnover,basedontheassumedturnoverrate(20%),biomassofﬁneroots(Table1),anddecompositionrateofﬁneroots(Table2andEq.(5)),showsthat−1thevalueswerebetween0.04and0.09gCm−2day.Thesevaluesaveragelessthan1.6%(0.5–5.0%)ofthetotalsoilrespirationrate.There-fore,wesuggestthatthisinﬂuenceisrelativelyminorinrelationtoourestimatesofheterotrophicandrootrespiration.

Atthisstudysite,thecontributionofrootrespira-tiontototalsoilrespirationinthegrowingseasonwasestimatedtobe27–71%,althoughthecontributionmightbemuchhigherinMayandNovember.Thesevaluesaresimilartothosereportedformanyforestecosystems(Hansonetal.,2000).However,coarserootsdirectlybelowtreestemswerenotincludedinourrootbiomassestimation.Therefore,thecontribu-tionofrootrespirationtostand-levelsoilrespirationmightbelargerthanourestimate.

ThepatternofseasonalchangeinRrwassome-whatdifferentfromthatinRh.Therootrespirationratewashighestinearlysummer(June)andthende-creased.Incontrast,nomarkedreductioninRhwasobservedinthesummerof2001.Thehighrootrespir-ationrateinearlysummermayhaveresultedfromthehighphysiologicalactivityassociatedwithrootgrowth(growthrespiration).TheperiodfrommidMaytotheendofJuneischaracterisedbyhighrootgrowthandturnover(EdwardsandHarris,1977;Joslin,1983).Hansonetal.(1993)foundthatthecontributionofRrtototalsoilrespirationcanchangedramaticallythroughoutayearasaresultofchangesinsoilres-pirationrateassociatedwithrootconstructioncosts.Ohashietal.(2000)andHögbergetal.(2001)sugges-tedthatrootgrowthrespiration,whichisassociatedwiththesynthesisofnewtissue,ﬂuctuatesirrespectiveofenvironmentalconditionssuchastemperature.Inourstudy,therewasasigniﬁcantrelationshipbetweenRhandsoiltemperature,whereasRrhadnosigniﬁc-antcorrelationwithsoiltemperature.ThisislargelyexplainedbythepeakrootrespirationinJune,whenleafexpansionwasactivebutsoiltemperaturewasstilllow.Infact,iftheJunedataareeliminated,therela-tionshipbetweenrootrespirationandsoiltemperatureissigniﬁcant(y=0.66e0.06x;P<0.05).

Severalresearchershaveproposedmodelsorequa-tionstopredictsoilrespirationratesfromabioticfactors,mainlytemperatureandsoilmoisture(e.g.,Bowdenetal.,1998;Koizumietal.,1999;Miel-

318

nickandDugas,2000).However,ourresultssuggestthatthefactorscontrollingtheseasonalchangeofres-pirationdifferbetweenthetwocomponentsofsoilrespiration(rootsandheterotrophicorganisms)andthatrootrespirationiscontrollednotonlybyabioticfactorsbutalsobyphenologicalchangesoftheplants.

Acknowledgements

WethankDr.WMooftheUniversityofTsukubaandDr.YHashimotoofObihiroUniversityforadviceandassistanceintheﬁeldmeasurements.WealsothankthemembersoftheInstituteforBasinEcosystemStudies,GifuUniversity,especiallyK.KurumadoandN.Miyamoto,andmembersoftheNakaneLaboratory,HiroshimaUniversity.

References

BowdenRD,BooneRD,MelilloJMandGarrisonJB1993Contributionsofabovegroundlitter,belowgroundlitter,androotrespirationtototalsoilrespirationinamixedhardwoodforest.Can.J.For.Res.23,1402–1407.

BowdenRD,KathleenMNandGinaMR1998Carbondioxideandmethaneﬂuxesbyaforestsoilunderlaboratory-controlledmoistureandtemperatureconditions.SoilBiol.Biochem.30,1591–1597.

ColemanDC1973Compartmentalanalysisof‘totalsoilrespira-tion’:anexploratorystudy.Oikos24,361–466.

EdwardsNT1991RootandsoilrespirationresponsestoozoneinPinustaedaL.seedlings.NewPhytol.118,315–321.

EdwardsNTandHarrisWF1977Carboninamixeddeciduousforestﬂoor.Ecology58,431–437.

EwelKC,CropperWPJrandGholzHL1987SoilCOFloridaslashplantations.II.Importanceofroot2evolutioninrespiration.Can.J.For.Res.17,330–333.

FaheyTJ,HughesJW,PuMandArthurMA1988Rootdecompos-itionandnutrientﬂuxfollowingwhole-treeharvestofnorthernhardwoodforest.For.Sci.34,744–768.

HansonPJ,WullschilegerSD,BohlmanSAandToddDE1993SeasonalandtopographicpatternsofforestﬂoorCOanuplandoakforest.TreePhysiol.13,1–15.

2efﬂuxfromHansonPJ,EdwardsNT,GratenCTandAndrewsJA2000Separatingrootandsoilmicrobialcontributionstosoilrespira-tion:areviewofmethodsandobservations.Biogeochemistry48,115–146.

HashimotoYandHyakumachiM1998Distributionofectomycor-rhizasandectomycorrhizalfungalinoculumwithsoildepthinabirchforest.J.For.Res.3,243–245

HögbergP,NordgrenA,BuchmannN,TaylorAFS,EkbladA,HögbergMN,NybergG,Ottosson-LofveniusMandReadDJ2001Large-scaleforestgirdlingshowsthatcurrentphotosyn-thesisdrivessoilrespiration.Nature411,7–792.

JoslinJD1983Thequantiﬁcationofﬁnerootturnoverinawhiteoakstand.PhDThesis,UniversityofMissouri,MO,USA.

KawamuraK,HashimotoY,SakaiTandAkiyamaT2001Effectsofphonologicalchangesofcanopyleafonthespatialandseasonal

variationsofunderstorylightenvironmentinacool-temperatedeciduousbroad-leavedforest.J.Jpn.For.Soc.83,231–237(inJapanese,withEnglishsummary).

KoizumiH,KontturiM,MarikoS,NakadaiT,BekkuYandMelaT1999Soilrespirationinthreesoiltypesinagriculturalecosys-temsinFinland.ActaAgric.Scand.Sec.BSoilPlantSci.49,65–74.

LeeM,NakaneK,NakatsuboT,MoWandKoizumiH2002EffectsofrainfalleventsonsoilCObroad-leavedforest.Ecol.Res.2ﬂuxinacool-temperatedeciduous17,401–409.

MalhiY,BaldocchiDDandJarvisPG1999Thecarbonbalanceoftropical,temperateandborealforests.PlantCellEnviron.22,715–740.

MarikoS,NishimuraN,MoW,MatsuiY,YokozawaM,SekikawaSandKoizumiH2000MeasurementofCO2ﬂuxesfromsoilandsnowsurfaceswithopendynamicchambertechnique.Environ.Sci.13,69–74.

MielnickPCandDugasWA2000SoilCO2ﬂuxintallgrassprairie.SoilBiol.Biochem.32,221–228.

NakaneK1978AmathematicalmodelofthebehaviorandverticaldistributionoforganiccarboninforestsoilsII.Arevisedmodeltakingthesupplyofrootlitterintoconsideration.Jpn.J.Ecol.28,169–177.

NakaneK,KohnoTandHorikoshiT1996Rootrespirationbeforeandjustafterclear-fellinginamaturedeciduous,broad-leavedforest.Ecol.Res.11,111–119.

OhashiM,GyokusenKandSaitoA2000Contributionofrootres-pirationtototalsoilrespirationinaJapanesecedar(CryptomeriajaponicaD.Don)artiﬁcialforest.Ecol.Res.15,323–333.

OlsonJS1963Energystorageandthebalanceofproducersanddecomposersinecologicalsystems.Ecology44,322–331.

RochettePandFlanaganLB1997Quantifyingrhizosphererespir-ationinacorncropunderﬁeldconditions.SoilSci.Soc.Am.J.61,466–474.

RochetteH,FlanaganLBandGregorichEG1999Separatingsoilrespirationintoplantandsoilcomponentsusinganalysisofnaturalabundanceofcarbon-13.SoilSci.Soc.Am.J.63,1207–1213.

SaigusaN,YamamotoS,MurayamaS,KondoHandNishimuraN2002Grossprimaryproductionandnetecosystemexchangeofacool-temperatedeciduousforestestimatedbytheeddycovariancemethod.Agric.For.Meteorol.112,203–215.

SaxeH,CannellMGR,JohnsenO,RyanMGandVourlitisG2001Treeandforestfunctioninginresponsetoglobalwarming.NewPhytol.149,369–400.

SinghJSandGuptaSR1977Plantdecompositionandsoilrespirationinterrestrialecosystems.Bot.Rev.43,449–528.TateKR,RossDJ,O’BrienBJandKelliherFM1993Carbonstorageandturnover,andrespiratoryactivity,inthelitterandsoilofanold-growthsouthernbeech(Nothofagus)forest.SoilBiol.Biochem.25,1601–1612.

ThierronVandLaudeloutH1996ContributionofrootrespirationtototalCO2efﬂuxfromthesoilofadeciduousforest.Can.J.For.Res.26,1142–1148.

UchidaM,NakatsuboT,HorikoshiTandNakaneK1998Contribu-tionofmicro-organismstothecarbondynamicsinblackspruce(Piceamariana)forestsoilinCanada.Ecol.Res.13,17–26.ValentiniR,MatteucciG,DolmanAJ,SchulzeED,RebmannC,MoorsEJ,GranierA,GrossP,JensenNO,PilegaardKetal.2000RespirationasthemaindeterminantofcarbonbalanceinEuropeanforests.Nature404,861–865.

YamamotoS,MurayamaS,SaigusaNandKondoH1999Sea-sonalandinter-annualvariationofCO2ﬂuxbetweenatemperateforestandtheatmosphereinJapan.Tellus51B,402–413.

因篇幅问题不能全部显示，请点此查看更多更全内容

查看全文

全部频道

生态