J Appl Ecol. 2019;56:2641–2651. wileyonlinelibrary.com/journal/jpe  | 2641© 2019 The Authors. Journal of Applied Ecology © 2019 British Ecological Society

Received:4March2019  |  Accepted:5August2019DOI: 10.1111/1365-2664.13498

R E S E A R C H A R T I C L E

Interaction between extreme weather events and mega‐dams increases tree mortality and alters functional status of Amazonian forests

Pamela Moser1  | Marcelo Fragomeni Simon2  | Marcelo Brilhante de Medeiros2  | Alexandre Bahia Gontijo3  | Flávia Regina Capelotto Costa4

1ProgramadePós‐GraduaçãoemEcologia,UniversidadedeBrasília,Brasília,Brazil2EmbrapaRecursosGenéticoseBiotecnologia,Brasília,Brazil3ServiçoFlorestalBrasileiro,Brasília,Brazil4InstitutoNacionaldePesquisasdaAmazônia,Manaus,Brazil

CorrespondencePamelaMoserEmail:moserpamela2@gmail.com

Funding informationConselhoNacionaldeDesenvolvimentoCientíficoeTecnológico;EnergiaSustentáveldoBrasilandFundaçãoArthurBernardes

HandlingEditor:CateMacinnis‐Ng

Abstract1. Forestresponsestochangesindroughtfrequencyisacriticalmatterforthefu-tureofAmazonforestsunderclimatechange,butequallyimportantisthemuchlessstudiedresponsetolargefloods,whichmayalsoincreasetreemortalityandchangeforestfunctionality.Further,forestvulnerabilitytofloodisbeingexacer-batedbylargehydroelectricdamsonAmazonriversthatputuplandenvironmentsnotadaptedtofloodatuniquerisk.

2. Toaddressthiscriticalknowledgegap,weevaluatedtheeffectsoftheextreme2014rainfallcoupledwiththenewlyconstructedJirauhydroelectricdamontreesurvivalandforestfunctionality,intheupperMadeiraRiverbasin.Weusedsur-veysofcampinaranawhite‐sandforests (stems>1cminseven1haplots)con-ductedbeforeandaftertheextremefloodtotesttrait‐basedecologicaltheorypredictionsoftheimpactoffloodonoverallcommunityfunction.

3. We found that flooding increased mortality by nearly five‐fold (from 3.2% to15.1%),mostlyinsmallertrees.Thislargemortalityinducedsignificantandcon-sistentshiftsincommunityfunction,towardsspecieswithconservativelifestrat-egies:directcomparisonof traitdifferencesbetweensurvivinganddying treesshowedthatsurvivorshadsmaller,highdensitystomata,andhigherleafdrymat-tercontent,wooddensityandroottissuedensity(RTD).Sizeanddensityofsto-mataandRTDwerethemostimportantpredictorsofspeciesmortalityrates.

4. Synthesis and applications.Althoughfocusedonasingleeventinonetypeofforest,thisworkhighlightsthegeneralimportance,andneedforfurtherstudy,ofinter-actionbetweenclimatechangeandmega‐damsinAmazonforests.Inparticular,weexpectthatcontinuedexpansionofhydroelectricdamsinAmazoniawilllikelyintensifythe impactof largefloodsonforestsmadenewlyvulnerablebythesedams,withsubstantialeffectonfutureforestfunctionalityinexpandedfloodplainareasacrosstheBasin.Hence,theseinteractionanalysesshouldberequiredintheBrazilianlegalinstrumentssuchastheenvironmentalimpactassessmentsanditsaccompanyingEnvironmentalImpactsReportsforlargeinfrastructureprojectsinAmazon.

2642  |    Journal of Applied Ecology MOSER Et al.

1  | INTRODUC TION

Climaticchange is increasing the frequencyofdroughts (Marengo&Espinoza,2016),butalsothefrequencyoflargefloodsintheAmazon(Barichivichetal.,2018;Ovandoetal.,2016).Theeffectsoflongerandmore frequent droughts are beingwidely studied, and reports haveshownanincreaseintreemortality(Condit,Hubbell,&Foster,1995;Flores et al., 2017), aswell as loss of biomass (Brienen et al., 2015;Phillips et al., 2009), in associationwith the large El Niño SouthernOscillationandTropicalNorthAtlanticeventsofthelasttwodecades(Marengo&Espinoza,2016).However,theeffectsoflargefloodshavereceived far less attention, despitebeing just as frequent andwide-spreadinthebasin,albeitgenerallyconfinedtothemarginsofrivers.Moreover,largehydroelectricdamshavebeenbuiltinmajorAmazonianriverswiththecorrespondingpotentialtoexacerbatethenegativeef-fectsofextremefloods(Finer&Jenkins,2012;Lees,Peres,Fearnside,Schneider,&Zuanon,2016).Recentresearchhasshownthatlong‐last-ingfloodsassociatedwithchangesinthehydrologicalregimecausedbylargedamshaveaffectedtreesurvivalwithlarge‐scaleimpactsonfloodplainvegetation(Assahiraetal.,2017;Resendeetal.,2019).

Theinteractionarisingfromthecouplingofclimatechangeandthe construction of large hydroelectric power dams has pushedfloodwaters towardupland sites (Ovandoet al., 2016)where thevegetationhasnotbeenevolutionarilyselected towithstandsuchpressure. Therefore, this study aimed to examine the effects ofthe2014floodonuplandsitesoftheupperMadeiraRiverbasininsouthern Amazonia where the Jirau hydroelectric mega‐damwasbuilt.Thisdambegantofillthereservoirin2013/2014,atimeframewhichcoincidedwithelevatedprecipitationintheAmazonbasininassociationwithwarmingoftheIndo‐Pacificandsubtropicalsouth-ernAtlanticOceans(Espinozaetal.,2014).Thisinteractioncausedarecord‐settingdischargeintheMadeiraRiver(58,000m3/s),risingtothehighest level (19.2m)everregistered inthehistoricalseries(ANA,2018),aspreviousrecordswere49,000m3/sand17.2m in1984 (Figure1a,b). This record‐setting rise inwater level resultedinfloodingalargeexpanseofuplandsfromBoliviatothesouthernBrazilian Amazon (Ovando et al., 2016), reaching terra‐firme siteswherelarge‐scalefloodshaveneverbeenregisteredbefore.

Weexaminedtheeffectsofthissevere2014floodontheup-landvegetationofcampinaranasoftheupperMadeiraRiverbasin.Campinaranas, alsoknownaswhite sand forests, areconsideredareservoirofendemicspecialistspecies(García‐Villacorta,Dexter,&Pennington,2016).ThisvegetationtypedominatestheNegroRiverbasin, but it also occurs in small patches across other Amazonianregions.The small patches and low fertilityofcampinaranasmakethisvegetationespecially susceptible todisturbancewith slow re-coveryrates(Adeney,Christensen,Vicentini,&Cohn‐haft,2016).In

thestudyarea,campinaranashaveanaturalseasonalfluctuationinwatertablelevels,whichcomeveryclosetothesurfaceduringtherainyseason,butthenretreattoadepthof4‐5minthedryseason.After the Jirau hydroelectric dam startedoperations in the upperMadeiraRiver,thewatertableundercampinaranasrosetoahigheraverageheightabovetypicalwatertablefluctuation(Figure1c,d).

Unusualfloodsareexpectedtoincreasetreemortality,accordingtospeciestraitsandtheintensityanddurationoftheflood(Colmer&Voesenek,2009;Wittmannetal.,2006).Inseasonallyfloodedforests,plantspecieshavebeenselectedforthisregularandpredictablestress.However,intheuplandenvironment,plantshavenotbeenselectedforit,andwe,therefore,expectedthatunnaturalflooding,suchasthatin2014,wouldfilterspeciesaccordingtotraitsrelatedto:hydraulics,asrepresentedbystomatasizeanddensity(SSandSD),xylemvesselareaanddensity(AandVD),vesselmeanhydraulicdiameter(Dmh),vesselsizetonumberratio (S)andwooddensity (WD),aswellasresourceacquisitionresponses,suchasspecificleafarea(SLA),leafdrymattercontent (LDMC),andspecificroot length (SRL).Moreover,plantsize(height,H) and thecapacity to conductoxygen in roots (root tissuedensity, RTD) are critical to survival. Flooding dramatically reducesoxygen levels in soils, leading toa switch to themuch lessefficientfermentativemetabolism(Parent,Capelli,Berger,Crèvecoeur,&Dat,2008).Insuchextremeconditions,absenceofoxygenandaccumula-tionoftoxicmetabolitesleadtothedeathoffineroots(Taiz&Zeiger,2006).Lowerenergyandimpairedrootsystemsthenleadtoimpairedsupplyofwaterandnutrientstotheplant.Theseconditionswouldbeparticularlystressfulforplantsofrapidgrowthwithcorrespondinglyhighdemandforwaterandnutrients,ashigherrespirationratesmayquicklyconsumeanystoredreserves.Thus,plantswithmoreconser-vativemetabolism(lowerSLAandSRL,buthigherLDMC;Cornelissenetal.,2003),saferhydraulicsystems(lowerWDandlowervesselandSS;Drakeetal.,2017;Poorteretal.,2010)andbettercapacitytoes-capefloodingstress(lowerRTDandhigherH)(Cornelissenetal.,2003;Ryser, Gill, & Byrne, 2011) should be better suited to survive afterfloodingcomparedtoplantswiththeoppositetraits.Thecapacityofrootstocontinuereceivingoxygenmaybecriticaltosurvivalinpro-longedflooding,andthismaydependonhavinglowertissuedensity,i.e.moreemptyspacesthroughwhichaircanmoreeasilyflow.Tallerplantswouldhaveobviousadvantagestoescapesomeofthestressesofflooding,suchastotalcoverageofleavesbywater.

Weexpectedthattheextremefloodswouldkilltreesselectively,accordingtospeciestraitsasdescribedabove,leadingtoachangeinthefunctionalcompositionoftheseforeststowardsmoreslow‐growingplants,which,ontheonehand,wouldmakecampinaranas more resilient in theeventofnew floods,but could,on theotherhand,decreaseitsshort‐termcapacityforcarbonsequestrationbe-sidesloweringforestrichnessanddiversity.Wehereinfollowedthe

K E Y W O R D S

dams,floodtolerance,functionaltraits,globalwarming,precipitationanomaly,roottissuedensity,stomatadensity,treemortality

     |  2643Journal of Applied EcologyMOSER Et al.

dynamics of these forests for 5 years, documenting themortalitylevels for thewhole tree community, by species and according tospeciestraits.Specifically,weasked(a)howtheextreme2014flood‐inducedmortalitywas related toplant size?; (b)whichplant traitswere related to increase inmortality rate for a range of 28mostabundantspecies?;and(c)howthecommunityfunctionalpropertieschangedaftertheextreme2014flooding?

2  | MATERIAL S AND METHODS

2.1 | Study area

ThestudywascarriedoutincampinaranasthatoccurintheareaofinfluenceoftheJirauhydroelectricdamlocatedintheupperMadeiraRiverbasin, southwesternAmazonia.Theclimate is tropicalhyper-thermalhumid(Cochrane&Cochrane,2010),withthehighestaverageannualtemperaturerangingfrom31.3to34.3°C,thelowestannualtemperaturerangingfrom19.3to22.8°Candmeanannualprecipita-tionbetween1,700and2,000mm(INMET,2019).Thepredominantvegetationisopenterra firmeforestgrowinginlatosols,whichoccupymost of theupland in the leftmarginof theMadeiraRiver (facing

downstream).Patchesofcampinaranasoccurontherightbankoftheriverinsilty,hydromorphicsoilthatissubjecttoseasonalwaterlog-gingcausedbytheelevationofthewatertable(Perigolo,Medeiros,&Simon,2017).Campinaranascomprisestuntedforeststhatvaryinthedensityofthetreestratum,rangingfromforestsupto20minheightand25%canopyopeningtoareaswithsparsetrees3–5mtallwith42%canopyopeningandadenseherbaceouslayerofpredominantlyCyperaceaefamily(Perigoloetal.,2017).

TheMadeiraisclassifiedasawhite‐waterriver,andundernatu-ralconditions,ithasfloodpulsesofapproximately10mhigh(Junketal.,2011;Figure1b).Becausetheriveroccurs inadeepcatchment,seasonalnaturalfloodingisrestrictedtoanarrowstripoflandonitsbanks.However, the fillingof theJirau reservoir,whichoccurredattheendof2013,modifiedtheriverfloodingregime,withanaverageincrease intheriver levelof4.2m intherainyseasonand5.2m inthedryseason(Figure1c). In2014,anextremefloodeventwasre-cordedinsouthwesternAmazonia,inassociationwiththeheatingofthePacific–IndianOceanandsubtropicalsouthernAtlantic(Espinozaetal.,2014),causinganintenseprecipitationinthemaintributariesoftheupperMadeiraRiverbasin.ThisfloodcoincidedwiththebeginningoftheoperationoftheJirauhydroelectricdamandboostedflooding

F I G U R E 1  HydrologicalregimeoftheupperMadeiraRiver.(a)Historicalseriesofdischargefrom1967to2017oftheMadeiraRiveratthePortoVelhoStation.(b)Historicalseriesofriverlevelfrom1967to2017(ANA,2018).Arrows.Dottedlinesin(a)and(b)representhistoricalaveragesofpeakdischargeandriverlevel(ANA,2018).(c)Fluctuationofthewaterlevelattwomonitoringstations(AbunãandMutum)nexttothestudyareaontheupperMadeiraRiverfrom2009to2017,showinganincreaseinbothminimumandmaximumlevelsaftertheoperationoftheJirauhydroelectricdam(Source:EnergiaSustentáveldoBrasil).(d)Watertablelevelinsevenforestplotsmeasuredwithpiezometers.Linesrepresentaveragespersite(AbunãandMutum)

2644  |    Journal of Applied Ecology MOSER Et al.

intheupperMadeiraRiver,whichreachedthehighestlevelobservedin the last fivedecades (Figure1b).The floodspreadmainlyon theflatterrainalongtherightbankoftheriver,floodinglargeexpansesofcampinaranas,aswellasterra firmeforestsandotheruplandenviron-ments.Thewatercolumnreachedamaximumof2.4mfromgroundinthecampinaranas,andthedurationofthefloodvariedacrossthesam-pledplotsbetween49and130daysasaconsequenceoftopographicvariation (TableS1 inAppendixS2). In lateryears, thecampinaranas werenotfloodedbytheoverflowoftheMadeiraRiver,butthewatertablebecameshallower in thedryseason,asa resultof theopera-tionoftheJirauhydroelectricdam(Figure1d).DescriptivestatisticsofdischargebeforeandafterthefillingoftheJiraudam,followingtheIndicators of Hydrologic Alteration method (Richter, Baumgartner,Powell, & Braun, 1996), are presented in Supporting Information(AppendixS1),andreinforcethepatternspresentedinFigure1.

2.2 | Field sampling

Wesampledseven1hapermanentplotsunderthevegetationmoni-toringprogrammecarriedoutintheareaaffectedbytheJiraureser-voir(Moseretal.,2014;Oliveira,Medeiros,Simon,Hay,&terSteege,2018;Perigoloetal.,2017).PlotswereallocatedfollowingRAPELD(Magnusson et al., 2005) sampling modules distributed systemati-callyalong theareaof influenceof theJirau reservoir,upstreamofthedam,andperpendiculartotheMadeiraRiver.Vegetationsampling(treeswithdiameteratbreastheight>1cm)wasperformedbiannuallybefore (2011and2013)andafter the fillingof the reservoir (2015)whendemographywasrecorded,includingmortalityrates.Wemeas-uredfunctionalattributesofthemostabundantspecies,representing>80%oftheabundanceperplotasinthe2011survey(Cornelissenetal.,2003).Wemeasured12functionalattributesofleaves,stemandrootsinFebruary2017,foratotalof271individuals(TablesS4andS5 inAppendixS2).Palmswerenot included.Brosimum acutifolium,Lacistema aggregatum,Meriania urceolata,Miconia prasina,Miconiasp.and Ternstroemia dentata, although within 80% of plot abundance,werenotsampledforfunctionalattributesbecauseofextremelyhighmortality rates after the2014 flood.Althoughmost traitmeasure-mentsweretakenafterflood(exceptplantheight),weassumethatthishadnoeffectinouranalysissincetraitsmeasuredshowedlimitedintraspecificvariation(seeTablesS4andS5inAppendixS2).

2.3 | Functional attributes

Weselected12functionalattributestodescribethefunctionalcom-positionofforestcommunitiesandunderstandtheimpactsofflood-ingaccordingtoplanttraits.Traitswererelatedtoresourceacquisitionresponses(leafandroot),hydraulics(vesselandwood)andplantsize(Fortunel,Ruelle,Beauchêne,Fine,&Baraloto,2014;Kramer‐Walteretal.,2016;Poorteretal.,2010;Wrightetal.,2004).Tomeasureleaf,stomataandxylemvesselsattributes,wecollectedabranchofapprox-imately1mforeachindividual.Immediatelyaftercollection,thebaseofthebranchwaswrappedinamoistpapertowelandconditionedinaplasticbagtomaintainhydrationuntilreachingthelaboratory.

The number of individuals collected for each attribute variedwithspecies.Thisvariationoccurredbecauseofthehighmortalityofsomespeciesafterthe2014flooding.ForSLA,LDMC,WD,SRLandRTD,thenumberofindividualsperspeciesrangedfrom1to30.Forstomataandxylemvessels, three individualsperspeciesweremeasured,exceptBrosimum rubescens and Parinarisp.,forwhichonlyoneindividualwasmeasured.

ThemeasuresofthefunctionalattributesfollowedtheprotocolofCornelissenetal.(2003).SLAwascalculatedbytheratiobetweenfreshleafareaanddrymass(mm2/mg).LDMCastheratiobetweentheleafdrymassandthewetmass(mg/g).Thenumberandlengthofstomatawereexaminedunderthemicroscopewith40×magnifi-cation.SRLwasobtainedbytheratiobetweenrootlengthandrootdrymass(m/g).RootswithhighRTDwascalculatedbytheratiobe-tweenrootdrymassandfreshvolume(g/cm3).

The Basic WD was obtained by the equation Db = [1/(Pf−Ps/Ps)+1/Gs](Smith,1954),wherePfistheweightofthesatu-ratedmass,Psisthedrymass,andGsisaconstantthatrepresentstheaveragedensityofthe“woodsubstance”equivalentto1.53g/cm3.Xylemvesselswecountedandmeasuredinamicroscopewith56×magnification.Fromthesemeasurements,wecalculatedmeanA(mm2);VD,givenbythenumberofvesselsperarea(n/mm2);meanhydraulicdiameterofthevessel[Dmh=(∑D4/n)1/4(mm2)],wheren isthenumberofvesselsandDisthevesseldiametercalculatedbyD=(4A/π)1/2(Scholz,Klepsch,Karimi,&Jansen,2013);andvesselsize‐to‐numberratio(S),givenbytherelationbetweensizeandnum-berofvessels(S = A/VD;Zanneetal.,2010).

Plantheightmean(H)wasobtainedbasedonthetotalnumberofindividualsperspeciesmeasuredinthe2011inventorywithatele-scopicmeasuringrodorlaserhypsometer.FurtherdetailsabouthowthefunctionalattributeswereobtainedareavailableinSupportingInformation(AppendixS2).

2.4 | Data analysis

Wecalculatedspeciesmortalityrates(Sheil,Burslem,&Alder,1995),basedoninventoriescarriedoutbefore(2011/2013)andafterthefillingof the reservoir (2013/2015). The variation in themortalityrategivenbythedifferencebetweenthemortalitybeforeandaftertheflood(Mort = M2013/2015–M2011/2013)wasusedasanestimateofincrementinmortalityasaresultofthe2014flood.

Statisticalcomparisonsofplantheightanddiameterbetweenin-dividualsthatdiedandsurvivedfloodingwereobtainedbytheMann–Whitneynon‐parametrictest.Therelationshipbetweenthefunctionalattributeswassummarizedwithaprincipalcomponentanalysis(PCA),using standardized variables for the relativeZ‐score scale.WeusedaGeneralized LinearModel (GLM) to test the relationship betweenmortalityrate(responsevariable)andfunctionalattributes(predictorvariables).Themodelwassetwithacombinationofpredictorspre‐se-lectedfromthosewithstrongerloadingsinthePCA,includingonepre-dictivevariable representingeach relevantplant component: LDMC(representingtheleaves),WD(wood),A(xylemvessels),SD(stomata),RTD(roots)andH(plantsize).

     |  2645Journal of Applied EcologyMOSER Et al.

To understand changes in the functional composition of thecommunityalongtime,wecalculatedtheplotcommunityweightedmean(CWM,theaverageofeachattributeweightedbytheabun-danceofeachspecies;Violleetal.,2007)ofeachattributeforeachcensustime.TheCWMwasobtainedwiththeaveragevalueofeachattributecollected in2017 foreachspeciesandweightedbyspe-ciesabundanceobtained in the2011,2013and2015 inventories,resultinginthreemeasurementsofCWMovertime,withFDpack-age(Laliberté,Legendre,&Shipley,2014).WeusedPCAtosumma-rizeandvisualizethetrajectoryofthecommunityineachplotbasedontheCWMcalculatedfor2011,2013and2015.Toevaluatetheeffectoffloodingonthecommunityfunctionalattributes,wealsocalculatedtherelativechangesofeachplotbeforeandafterflood-ing,whereΔ=[(CWM2015 – CWM2011)/CWM2011]×100(Carreño‐Rocabadoetal.,2012).DifferencesbetweenCWM2011andCWM2015 weretestedwithapairedt test.FortheCWMcalculations,all28speciesrepresenting80%ofthecommunity'sabundancewereused.InthePCAanalysisandGLMmodel,only25specieswereconsid-eredsincewechosenottoincludethethreespecieswiththelowestabundancein2011(≤10individuals),thusavoidingabiasinthemor-talityratethatmayhaveoccurredasaresultofrandomprocessesnotrelatedtoflooding.AllanalyseswereperformedinRplatform(RCoreTeam,2018).

3  | RESULTS

3.1 | Tree mortality

Plots flooded by the rising waters of the upperMadeira River in2013/2014 had highly increased annual mortality rates after theflood (average 10.5%) compared to the previous period (1.9%;Figure 2a; Table S1 in Appendix S2). Plants that died after the

floodwere smaller,with lower diameter (U = 1,274,100,p < .001; Figure2b)andheight(U=1,429,800,p<.001;Figure2c)thansurvi-vors.Floodingreachedamaximumheightof2.4mfromthegroundinsomesites,asmeasuredbythefloodmarksontreetrunks,andindividualsbelowthissizehadalltheirleavessubmerged.Althoughfloodintensityvariedbetweenplots (TableS1 inAppendixS2),wefoundno correlationbetween increment inmortality rate and thenumberoffloodeddays(p>.05).

3.2 | Differential mortality across species

Atthespecieslevel,largedifferencesintolerancetofloodingwereobserved.Iryanthera juruensis,Clusiasp.andLicania caudatasufferedhighmortality(>38%),while12specieshadlowmortality(<1%),in-cludingRuizterania retusa,whichisdominantinthestudiedforestswithalmostathirdofallindividuals(TableS2inAppendixS2).

3.3 | Relationship among the functional traits

TheaxisPCA1describedagradientthatwaspositivelycorrelatedwith theS,H, SRL,A and Dmh and thatwas negatively correlatedwiththeVD.TheaxisPCA2describedagradientthatwaspositivelycorrelatedwith theLDMC,WDandRTDand thatwasnegativelycorrelatedwiththeSLAandSS(Figure3;TableS3inAppendixS2).Descriptivestatisticsofthe12functionaltraitsforthe28speciesareprovidedinTablesS4andS5inAppendixS2.

3.4 | Best predictors of mortality

Althoughseveraltraitswereassociatedwiththedifferentialmortal-ityof species, ageneral linearmodel retainedSDandRTDas themostimportantpredictors(Table1);mortalityincreaseswithlowSD

F I G U R E 2  TreemortalityinsevenforestplotsintheupperMadeiraRiverbasin,southernAmazonia.(a)Averagemortalityratesbefore(2011–2013)andafterthe2014recordflood(2013–2015)insevenforestplots.(b)Averagediameteratbreastheight(DBH)and(c)heightoftreesthatsurvivedordiedafter2014extremeflood(N=4,920).U=Mann–Whitneynon‐parametrictest

2646  |    Journal of Applied Ecology MOSER Et al.

andRTD. SD is negatively correlatedwith SS (r = −.56), and bothtraitsstronglyaffectstomatalconductanceandthereforephotosyn-thesis(Farquhar&Sharkey,1982).

3.5 | Community level changes

Theincreasedselectivemortalityofspeciesresultingfromtheex-treme2014floodledtoachangeinthefunctionalcompositionof

tree community. This change encompasses a significant increasein theCWMsof leafdrymattercontent,WD,vessel size tonum-berratio,SDandplantheight,aswellasadecreaseinSS(Figure4).Thisrepresentedadirectionalchangeinthefunctionaltrajectoryoftheseforeststowardsamoreconservativefunctionalcomposition(Figure5;TableS4inAppendixS2).

4  | DISCUSSION

TheinteractionbetweenthenewJiraumega‐damandtheabove‐aver-ageprecipitationanomaly in2014resulted infloodwaterscoveringareasnotpreviouslysubjectedtoinundationbeforedamconstruction,andalong‐lastingflooding(49–130days;TableS1inAppendixS2).Inthisstudy,wehaveshownthatthismega‐floodcausedasignificantin-creaseintreemortalityratesinthecampinaranasoftheupperMadeiraRiverbasinandthatthismortalitywasselective,affectingmostlytreeswithlowerheight,acquisitiveresourcestrategiesandpotentiallylowerstomatalcontrol.Thesurvivingcommunitychangeddirectionallyto-wardsconservativestrategies.

TreemortalityinducedbytheinteractionbetweentheJiraures-ervoirandtheextremerainfalleventin2014was5timeshigher(upto15.1%)thanthatobservedinnaturalconditionsandnormalyears(up to3.2%).Backgroundmortality rates in tropical rainforestsarearound2%peryearwhilemoratiltyrates>5%peryearareconsideredcatastrophic(Lugo&Scatena,1996).Themortalitylevelfoundhereiscomparabletothatobservedundertherecordwindthrowsof2005in centralAmazonia,which increasedmortality by 23% and,whenscaled to the landscape, included0.32milliondead trees (Negrón‐Juárezetal.,2010).Effectsofdroughtshavebeenvariable,dependingondroughtintensityandregion,beingaroundtwotimeshigherthanbackground levels across theAmazon (McDowell et al., 2018) andcentralAmazonia(Williamsonetal.,2000).Thesecomparisonsshowthatmortalitylevelsbyextremerainfalleventsthatdrivefloodinginhuman‐modifiedcontexts,suchashydroelectricreservoirs,are justasdeleterious,ormoreso,thanotherclimate‐inducedchangesintheAmazonandhavethepotentialtochangethelandscapeofrivermar-gins.Althoughlargelyconfinedtothemarginsofrivers,theupscalingoftheseeffectstothebasinmayreachnon‐trivialvalues,astheareacoveredbyriverinevegetationisestimatedtobe12%ofthesevenmillion square kilometers of the Amazon basin (Junk et al., 2011).Forexample,thecombinedeffectsof2014extremerainfallandtheinstallationofthetwomegahydroelectricsJirauandSantoAntônio(locatedca.100kmdownstreamfromJirau)resultedinanestimatedfloodedareaofmorethan800km2acrosstheupperMadeiraRiverbasin(Cochrane,Matricardi,Numata,&Lefebvre,2017).

Although subject to seasonal flooding by groundwater fluc-tuation, campinaranas in the study area have not evolved underthepressureofriverinundations.Consequently,speciesarenotasmorphologicallyorphysiologicallyadaptedtothisstressasplantsthatevolvedintherivermarginssubjectedtopredictableannualriver floods. Under the unexpected long‐standing 2014 flood,plant species with acquisitive strategies associated with higher

F I G U R E 3  Principalcomponentanalysisof12functionaltraitsof25treespeciesaffectedbytheextreme2014floodintheupperMadeiraRiverbasin,southernAmazonia.Contributionofeachvariabletotheordinationisdisplayedproportionaltoarrowsize.A,meanvesselarea;Dmh,vesselmeanhydraulicdiameter;H,height;LDMC,leafdrymattercontent;RTD,roottissuedensity;S,vesselsizetonumberratio;SD,stomatadensity;SLA,specificleafarea;SRL,specificrootlength;SS,stomatasize;VD,vesseldensity;WD,wooddensity

TA B L E 1  Generallinearmodeldescribingtheeffectoftraitsonthevariationofmortalityrateof25treespeciesaffectedbythe2014extremefloodintheupperMadeiraRiverbasin,southernAmazonia

Estimate SE t Value P

(Intercept) 10.313 2.960 3.484 .003

Leafdrymattercontent

−4.596 4.765 −0.965 .349

Wooddensity 1.836 4.468 0.411 .687

Vesselarea −0.497 3.377 −0.147 .885

Stomatadensity

−8.350 3.438 −2.429 .027

Roottissuedensity

−8.625 3.420 −2.522 .022

Treeheight −1.978 3.722 −0.532 .602

     |  2647Journal of Applied EcologyMOSER Et al.

energetic demands that become too costly in the new floodedconditionswereespeciallyaffectedandhad increasedmortality.The oxygen reduction in flooded soils leads to a change to themuchlessefficientfermentativemetabolism(Parentetal.,2008).In this condition, carbohydrate reserves may be quickly used,andtheplantmayenter intocarbonstarvation.Soilhypoxiaandanoxia reduce root permeability and hydraulic conductivity innon‐tolerantspecies,actuallycausingsymptomssimilar to thoseseenindroughtconditions(Parentetal.,2008).Acquisitiveplantsalsohave largertranspirationsurfaces,giventhehigher leafsur-face‐to‐mass relationship, which can increase the stress causedbydecreasedwaterconductancefromroots.Underthesecircum-stances, acquisitive plants exhaust their carbohydrate reserves,water absorptionby roots is inhibited, and transpiration is high,while their large stomata arenot efficient to controlwater loss.

Thesemultiplestressesmaykillplantsbybothcarbonstarvationandhydraulicfailure(McDowelletal.,2008).

Plantheightwasalsoimportanttodetermineplantfateundertheextreme2014flood.Greaterheightsmayhelpplantsescapeflood-ingandincreasesurvival,sincesubmersionofthecanopyrestrictscontactofleaveswithoxygenandlight(Parolin&Wittmann,2010),decreasingorceasinggasexchangeandphotosynthesis (Colmer&Voesenek,2009).Plantspecies thatevolvedunderperiodic flood-ing are adapted to submergence (Fernández, 2006), but not terra firme or campinaranaspecies.Smallplantsthatweresubmergedincampinaranas experienced the stresses of soil anoxia, overlappedwith thestressof leaf submergenceandwere those thatsufferedthemost,withhighmortalitylevels.

SizeanddensityofstomataandRTDwerethemostimportantpredictors of species mortality rates (Figure S1 in Appendix S2).

F I G U R E 4  Relativechangeinthecommunityweightedmean(CWM)oftraitsinsevenforestplots(andaveragedacrossallplots,horizonalline)associatedwiththe2014recordfloodintheupperMadeiraRiverbasin,southernAmazon.BarsrepresentthepercentdifferenceineachCWMtraitineachplotbeforeandaftertheflood;thestatisticalsignificanceistestedusingpairedttestacrossallplots.A,meanvesselarea;Dmh,vesselmeanhydraulicdiameter;H,height;LDMC,leafdrymattercontent;RTD,roottissuedensity;S,vesselsizetonumberratio;SD,stomatadensity;SLA,specificleafarea;SRL,specificrootlength;SS,stomatasize;VD,vesseldensity;WD,wooddensity

2648  |    Journal of Applied Ecology MOSER Et al.

Small stomatacanopenandclose faster,given thehigherpropor-tionofcellmembranetovolume(Drakeetal.,2017),makingitpos-sible to better adjust to unfavorable conditions. Plantswith goodstomatalcontrolcanreducestomatalconductanceupto30%underexperimental flooding (Lopez & Kursar, 1999) in response to theloweredwaterabsorptionbyrootsinanoxicsoils(Pezeshki,1993).Lowoxygensupplyandaccumulationoftoxiccompoundsinfloodedsoilsdeterioraterootsofnon‐adaptedplants,loweringrootgrowthand,inworstcases,causerotting,thusopeningthedoortoinvasionofpathogens(Kozlowski,1997).Ingeneral,thinandlightroots(i.e.lowerRTD) aremore susceptible to these conditions, and specieswiththistraithadhighermortalityrates.

Atthecommunitylevel,theselectivemortalityimposedbytheextremefloodin2014filteredoutacquisitivespecies,directionallychangingthefunctionalstructuretowardsconservativestrategies,aphenomenonthatultimatelyledtochangesinthefunctioningofthisforestecosystem.Amoreconservativetreecommunitymaybemoreresistanttonewfloodeventsinthefuture,butwillhavealowercarbonsequestrationcapacityintheshorttomediumterm.Ifnewfloodsarenotcontrolledandstrikeagain inthesameregions, thenew community/functional composition and carbon sequestrationpatternmaybecomeapermanentstate,aspredictedbyalternativestatestheory.Floresetal.(2017)suggestedthatfloodedforestsmaybethemostvulnerableforestintheAmazonduetothesynergismofdroughtsandfire.Buthereweshowedthatforestsinthemarginsofrivers,notsubjecttoregularflooding,arealsovulnerabletosyn-ergiesofclimatechangeandlargeinfrastructureprojects,thatmaylead these ecosystems to irreversible tipping points. Additionally,thelossofsensitivespeciesreducesthefunctionaldiversityoftheforest,whichmayleadtodecreasedecosystemservices,suchasre-sourcestothefauna,withstillunsuspectedcascadingeffects.

ThenegativeeffectsoffloodingonAmazonianecosystemshaveincreasedinthelastdecadesowingtoincreasedconstructionofhy-droelectricdams(Finer&Jenkins,2012).In2013/2014,thefillingof

the Jirau hydroelectric dam reservoir coincided with precipitationanomalies in the upperMadeiraRiver basin that exceeded averagerainfallbyaround100%(Espinozaetal.,2014).Suchunprecedentedrainfall was associated with the warming of the Indo‐Pacific andwithexceptionallywarmconditions in subtropical southernAtlanticOceans, which, in turn, increased the transport of humidity oversouthwestern Amazonia (Espinoza et al., 2014). The walls of Jirauand SantoAntônio hydroelectrics dammed the excess precipitationin2014,andthefloodspreadovermorethan800km2,coveringanarea 64.5% larger than predicted by environmental impact studies(Cochraneetal.,2017).This floodalso lastedmuch longer thanthetypicalannual floods,stressingeventhevarzea (seasonally flooded)forests.Theinteractionbetweentheexpandedconstructionofhydro-electricdamsandtheincreasedfrequencyofextremeclimaticeventsowingtoclimatechange(Barichivichetal.,2018;Marengo&Espinoza,2016)islikelytomaketheconditionsthatproducemega‐floods,suchthatin2014,amorefrequentoccurrenceintheAmazon.Theincreaseinfrequencyanddurationoffloodsimpliesachangeintheannualriverfloodpattern,affectingthedynamicsofforestsintherivermargins,withincreasedmortalityandalteredfunctionality.Ourfindingssug-gestthatdecisionsonthebuildingofnewdamsandalsotheoperationofthosealreadyinplace(eg.Resendeetal.,2019)shouldbere‐evalu-atedinthelightofthedramaticnegativeeffectsdemonstratedhere.

Ourresultssuggestthattreemortalitycausedbythecombina-tionofextremerainfall/riverdischargeassociatedwithdammingis substantially higher than average mortality in floodplain for-ests(eg.Homeier,Kurzatkowski,&Leuschner,2017;Nebel,Kvist,Vanclay,&Vidaurre, 2001). Evenduring unusually severe flood-ingseasons,treemortalityintropicalriparianforestsseemstobelowerthandetectedhere,probablybecauseflooding intensity iswithinspeciesfloodingtolerance.Forexample,mortalityratesas-sociatedwith extreme floods in riparian forests in the ParaguaiRiver(4.1%peryear;Damasceno‐Junior,Semir,Santos,&Leitão‐Filho,2004)andinsoutheasternBrazil(2.2%peryear;Silva,Berg,

F I G U R E 5  Functionaltrajectoriesofsevenforestplotsfrombefore(2011,2013,arrowtail)toafter(2015,arrowhead)the2014recordfloodintheupperMadeiraRiverbasin,southernAmazonia.Eachdatapointcorrespondstoprincipalcomponentanalysis(PCA)scoresbasedon12communityweightedmean(CWM)traits.FunctionaltraitsthatcontributedmosttoeachPCAaxisaredisplayed:RTD,roottissuedensity;S,vesselsizetonumberratio;SLA,specificleafarea;andWD,wooddensity

     |  2649Journal of Applied EcologyMOSER Et al.

Higuchi,&Nunes,2011)suggestthatnaturalfloodsaloneareun-likely to generate levels of mortality as high as reported in ourstudysite.

Intropicalforests,treemortalityhasincreasedinrecentdecadesas a consequence of severe droughts, competitionwith lianas, firesandwindstorms,withimplicationsforbothbiodiversityconservationandmaintenanceofthecarboncycle(McDowelletal.,2018).Inlinewith recent studies (Assahira et al., 2017;Resendeet al., 2019),wehaveshownthatfloodinganomaliesarealsoimportantdriversoftreemortality in theAmazon,with effects that are comparable toothermajordriverssuchasdroughtandwindstorms.Theinteractioneffectbetweenextremeclimaticeventsandtheconstructionofhydroelec-tricdamsislikelytoincreasethefrequencyofmajorfloods,leadingtohighertreemortalityandchangesinthefunctionalcompositionoffor-estcommunities.Hence,theseinteractionanalysesshouldberequiredintheBrazilianlegalinstrumentssuchastheenvironmentalimpactas-sessmentsanditsaccompanyingEnvironmental ImpactsReportsforlargeinfrastructureprojectsinAmazon.

ACKNOWLEDG EMENTS

WethankValdeciGomes,GlocimarSilva,ValdemarSilva,WellyngtonEspindola, Iris Kraievski, Caio Silva, Antônio dos Santos, SérgioNoronha,LuisPalhares,WashingtonOliveira,BrunoWalter,EduardaDias, alongwithnumerous technicians and studentswho contrib-utedtobothfieldandherbariumwork.WearegratefultoMercedesBustamante laboratory (UniversityofBrasilia)and theLaboratóriodeProdutosFlorestais(LPF‐SFB)staffforassistancewithrootandxylemanalysis,respectively.Thisarticleispartoffirstauthor'sdoc-toral thesis (UniversityofBrasília)with support fromtheNationalCouncil of Technological and Scientific Development (CNPq).WethankEnergiaSustentáveldoBrasilandFundaçãoArthurBernardesforfinancialsupport.

AUTHORS' CONTRIBUTIONS

P.M.,M.B.d.M.,M.F.S.andF.R.C.C.designedthestudy,P.M.,M.F.S.,andF.R.C.C.analyseddatawithinputfromM.B.d.M.andA.B.G.P.M.,M.F.S.andF.R.C.C.wrotethepaper,withinputfromM.B.d.M.andA.B.G.

DATA AVAIL ABILIT Y S TATEMENT

DataavailableviatheDryadDigitalRepositoryhttps://doi.org/10.5061/dryad.8041f4r(Moser,Simon,Medeiros,Gontijo,&Costa,2019).

ORCID

Pamela Moser https://orcid.org/0000‐0003‐1956‐9742

Marcelo Fragomeni Simon https://orcid.org/0000‐0002‐5732‐1716

Marcelo Brilhante Medeiros https://orcid.org/0000‐0001‐7619‐6001

Alexandre Bahia Gontijo https://orcid.org/0000‐0003‐2728‐0383

Flávia Regina Capelotto Costa https://orcid.org/0000‐0002‐9600‐4625

R E FE R E N C E S

Adeney,J.M.,Christensen,N.L.,Vicentini,A.,&Cohn‐Haft,M.(2016).White‐sandecosystems inAmazonia.Biotropica,48, 7–23.https://doi.org/10.1111/btp.12293

ANA. (2018). Agência Nacional de Águas. HIDROWEB: DadosHidrológicos. Retrieved from www.snirh.gov.br/hidroweb/publico/medicoes_historicas_abas.jsf

Assahira,C.,Piedade,M.T.F.,Trumbore,S.E.,Wittmann,F.,Cintra,B.B.L.,Batista,E.S.,…Schöngart,J.(2017).Treemortalityofaflood‐adaptedspeciesinresponseofhydrographicchangescausedbyanAmazonianriverdam.Forest Ecology and Management,396,113–123.https://doi.org/10.1016/j.foreco.2017.04.016

Barichivich, J., Gloor, E., Peylin, P., Brienen, R. J. W., Schöngart, J.,Espinoza,J.C.,&Pattnayak,K.C. (2018).Recent intensificationofAmazonfloodingextremesdrivenbystrengthenedWalkercircula-tion.Science Advances,4,eaat8785.https://doi.org/10.1126/sciadv.aat8785

Brienen, R. J.W., Phillips,O. L., Feldpausch, T. R., Gloor, E., Baker, T.R.,Lloyd,J.,…Zagt,R.J. (2015).Long‐termdeclineoftheAmazoncarbon sink. Nature, 519, 344–348. https://doi.org/10.1038/nature14283

Carreño‐Rocabado,G.,Peña‐Claros,M.,Bongers,F.,Alarcón,A.,Licona,J.C.,&Poorter,L.(2012).Effectsofdisturbanceintensityonspeciesandfunctionaldiversity inatropicalforest.Journal of Ecology,100,1453–1463.https://doi.org/10.1111/j.1365‐2745.2012.02015.x

Cochrane, S.M. V.,Matricardi, E. A. T., Numata, I., & Lefebvre, P. A.(2017). Landsatbased analysis of mega dam flooding impacts inthe Amazon compared to associated environmental impact as-sessments: Upper Madeira River example 2006–2015. Remote Sensing Applications: Society and Environment, 7, 1–8. https://doi.org/10.1016/j.rsase.2017.04.005

Cochrane,T.T.,&Cochrane,T.A.(2010).Amazon Forest & Savanna Lands: A guide to the climates, vegetation, landscapes and soils of central tropi‐cal South America.ScottsValley,CA:CreateSpace.

Colmer,T.D.,&Voesenek,L.A.C.J.(2009).Floodingtolerance:Suitesofplanttraitsinvariableenvironments.Functional Plant Biology,36,665–681.https://doi.org/10.1071/FP09144

Condit, R., Hubbell, S. P., & Foster, R. B. (1995). Mortality rates of205 Neotropical tree and shrub species and the impact of a se-vere drought. Ecological Monographs, 65, 419–439. https://doi.org/10.2307/2963497

Cornelissen, J. H. C., Lavorel, S., Garnier, E., Díaz, S., Buchmann, N.,Gurvich,D. E.,… Poorter,H. (2003). A handbook of protocols forstandardised and easy measurement of plant functional traitsworldwide. Australian Journal of Botany, 51, 335–380. https://doi.org/10.1071/BT02124

Damasceno‐Junior,G.A., Semir, J., Santos,F.A.M.D.,&Leitão‐Filho,H.D.F. (2004).Treemortality in a riparian forest atRioParaguai,Pantanal,Brazil,afteranextremeflooding.Acta Botanica Brasilica,18,839–846.https://doi.org/10.1590/S0102‐33062004000400014

Drake, J. E., Power, S. A., Duursma, R. A.,Medlyn, B. E., Aspinwall,M. J., Choat, B., … Tissue, D. T. (2017). Stomatal and non‐sto-matal limitations of photosynthesis for four tree species underdrought: A comparison of model formulations. Agricultural and Forest Meteorology,247,454–466.https://doi.org/10.1016/j.agrformet.2017.08.026

Espinoza,J.C.,Marengo,J.A.,Ronchail, J.,Carpio,J.M.,Flores,L.N.,& Guyot, J. L. (2014). The extreme 2014 flood in south‐westernAmazon basin: The role of tropical‐subtropical SouthAtlantic SSTgradient. Environmental Research Letters, 9, 124007. https://doi.org/10.1088/1748‐9326/9/12/124007

Farquhar,G.D.,&Sharkey,T.D.(1982).Stomatalconductanceandpho-tosynthesis.Annual Review of Plant Physiology,33,317–345.https://doi.org/10.1146/annurev.pp.33.060182.001533

2650  |    Journal of Applied Ecology MOSER Et al.

Fernández,M.D.(2006).Changesinphotosynthesisandfluorescenceinresponsetoflooding inemergedandsubmerged leavesofPouteria orinocoensis. Photosynthetica, 44, 32–38. https://doi.org/10.1007/s11099‐005‐0155‐2

Finer, M., & Jenkins, C. N. (2012). Proliferation of hydroelectricdamsintheAndeanAmazonandimplicationsforAndes‐Amazonconnectivity. PLoS ONE, 7, 1–9. https://doi.org/10.1371/journal.pone.0035126

Flores, B. M., Holmgren, M., Xu, C., van Nes, E. H., Jakovac, C. C.,Mesquita,R.C.G.,&Scheffer,M.(2017).FloodplainsasanAchilles'heel of Amazonian forest resilience. Proceedings of the National Academy of Sciences of the United States of America,114,4442–4446.https://doi.org/10.1073/pnas.1617988114

Fortunel,C.,Ruelle,J.,Beauchêne,J.,Fine,P.V.A.,&Baraloto,C.(2014).Wood specific gravity and anatomy of branches and roots in 113Amazonian rainforest tree speciesacrossenvironmentalgradients.New Phytologist,202,79–94.https://doi.org/10.1111/nph.12632

García‐Villacorta,R.,Dexter,K.G.,&Pennington,T.(2016).Amazonianwhite‐sandforestsshowstrongfloristiclinkswithsurroundingoligo-trophichabitatsandtheGuianaShield.Biotropica,48,47–57.https://doi.org/10.1111/btp.12302

Homeier,J.,Kurzatkowski,D.,&Leuschner,C.(2017).Standdynamicsofthedrought‐affectedfloodplainforestsofAraguaiaRiver,BrazilianAmazon. Forest Ecosystems, 4, 1–10. https://doi.org/10.1186/s40663‐017‐0097‐8

INMET. (2019). Instituto Nacional de Meteorologia. Retrieved fromhttp://www.inmet.gov.br/portal/index.php?r=clima/normaisclimatologicas

Junk,W.J.,Piedade,M.T.F.,Schöngart,J.,Cohn‐Haft,M.,Adeney,J.M.,&Wittmann,F.(2011).Aclassificationofmajornaturally‐occurringAmazonian lowland wetlands.Wetlands, 31, 623–640. https://doi.org/10.1007/s13157‐011‐0190‐7

Kozlowski, T. T. (1997). Responses of woody plants to flooding andsalinity. Tree Physiology, 17, 490. https://doi.org/10.1093/treephys/17.7.490

Kramer‐Walter, K. R., Bellingham, P. J., Millar, T. R., Smissen, R. D.,Richardson,S.J.,&Laughlin,D.C.(2016).Roottraitsaremultidimen-sional:Specificroot lengthis independentfromroottissuedensityand the plant economic spectrum. Journal of Ecology, 104, 1299–1310.https://doi.org/10.1111/1365‐2745.12562

Laliberté,E.,Legendre,P.,&Shipley,B.(2014).FD:Measuringfunctionaldiversityfrommultipletraits,andothertoolsforfunctionalecology.Rpackageversion1.0‐12..

Lees,A.C.,Peres,C.A., Fearnside,P.M., Schneider,M.,&Zuanon, J.A.S.(2016).HydropowerandthefutureofAmazonianbiodiversity.Biodiversity and Conservation,25,451–466.https://doi.org/10.1007/s10531‐016‐1072

Lopez,O.R.,&Kursar,T.A.(1999).Floodtoleranceoffourtropicaltreespecies.Tree Physiology,19,925–932.https://doi.org/10.1093/treephys/19.14.925

Lugo,A.,&Scatena,F.(1996).Backgroundandcatastrophictreemortal-ity intropicalmoist,wet,andrainforests.Biotropica,28,585–599.https://doi.org/10.2307/2389099

Magnusson,W.E.,Lima,A.P.,Luizão,R.,Luizão,F.,Costa,F.R.C.,deCastilho,C.V.,&Kinupp,V. F. (2005).RAPELD:AmodificationoftheGentrymethodforbiodiversitysurveysinlong‐termecologicalresearch sites.Biota Neotropica,5, 21–26.https://doi.org/10.1590/S1676‐06032005000300002

Marengo,J.A.,&Espinoza,J.C.(2016).ExtremeseasonaldroughtsandfloodsinAmazonia:Causes,trendsandimpacts.International Journal of Climatology,36,1033–1050.https://doi.org/10.1002/joc.4420

McDowell,N.,Allen,C.D.,Anderson‐Teixeira,K.,Brando,P.,Brienen,R.,Chambers,J.,…Xu,X.(2018).Driversandmechanismsoftreemor-talityinmoisttropicalforests.New Phytologist,219,851–869.https://doi.org/10.1111/nph.15027

McDowell,N.,Pockman,W.T.,Allen,C.D.,Breshears,D.D.,Cobb,N.,Kolb, T., … Yepez, E. A. (2008).Mechanisms of plant survival andmortality during drought:Why do some plants survivewhile oth-erssuccumbtodrought?New Phytologist,178,719–739.https://doi.org/10.1111/j.1469‐8137.2008.02436.x

Moser,P.,Oliveira,W.L.,Medeiros,M.B.,Pinto,J.R.,Eisenlohr,P.V.,Lima,I.L.,…Simon,M.F.(2014).Treespeciesdistributionalongenvironmen-tal gradients in an area affected by a hydroelectric dam in southernAmazonia.Biotropica,46,367–376.https://doi.org/10.1111/btp.12111

Moser,P.,Simon,M.F.,Medeiros,M.B.,Gontijo,A.B.,&Costa,F.R.C.(2019).Datafrom:Interactionbetweenextremeweathereventsandmega‐damsincreasestreemortalityandaltersfunctionalstatusofAmazonianforests.Dryad Digital Respository,https://doi.org/10.5061/dryad.8041f4r

Nebel,G.,Kvist,L.P.,Vanclay,J.K.,&Vidaurre,H.(2001).ForestdynamicsinfloodplainforestsinthePeruvianAmazon:Effectsofdisturbanceand implications formanagement.Forest Ecology and Management,150,79–92.https://doi.org/10.1016/S0378‐1127(00)00682‐4

Negrón‐Juárez,R.I.,Chambers,J.Q.,Guimaraes,G.,Zeng,H.,Raupp,C.F.M.,Marra,D.M.,…Higuchi,N.(2010).WidespreadAmazonforesttreemortalityfromasinglecross‐basinsqualllineevent.Geophysical Research Letters,37,1–5.https://doi.org/10.1029/2010GL043733

Oliveira,W.L.,Medeiros,M.B.,Simon,M.F.,Hay,J.,&terSteege,H.(2018).Theroleofrecruitmentanddispersallimitationintreecom-munity assembly inAmazonian forests.Plant Ecology and Diversity,11,1–12.https://doi.org/10.1080/17550874.2018.1474960

Ovando, A., Tomasella, J., Rodriguez, D. A.,Martinez, J.M., Siqueira‐Junior,J.L.,Pinto,G.,…vonRandow,C.(2016).Extremefloodeventsin the Bolivian Amazon wetlands. Journal of Hydrology: Regional Studies,5,293–308.https://doi.org/10.1016/j.ejrh.2015.11.004

Parent,C.,Capelli,N.,Berger,A.,Crèvecoeur,M.,&Dat,J.F. (2008).Anoverviewofplantresponsestosoilwaterlogging.Plant Stress,2,20–27.

Parolin,P.,&Wittmann,F.(2010).Struggleintheflood:Treeresponsesto flooding stress in four tropical floodplain systems. AoB Plants,2010,plq003.https://doi.org/10.1093/aobpla/plq003

Perigolo,N.A.,deMedeiros,M.B.,&Simon,M.F. (2017).Vegetationtypesof theupperMadeiraRiver inRondônia,Brazil.Brittonia,69,423–446.https://doi.org/10.1007/s12228‐017‐9505‐1

Pezeshki, S. R. (1993). Differences in patterns of photosynthetic re-sponsestohypoxiainflood‐tolerantandflood‐sensitivetreespecies.Photosynthetica,28,423–430.

Phillips, O. L., Aragao, L. E. O. C., Lewis, S. L., Fisher, J. B., Lloyd, J.,Lopez‐Gonzalez,G.,…Torres‐Lezama,A. (2009).Droughtsensitiv-ity of the amazon rainforest.Science,323, 1344–1347. https://doi.org/10.1126/science.1164033

Poorter, L.,McDonald, I., Alarcón, A., Fichtler, E., Licona, J.‐C., Peña‐Claros, M., … Sass‐Klaassen, U. (2010). The importance of woodtraits andhydraulic conductance for theperformanceand lifehis-tory strategies of 42 rainforest tree species.New Phytologist,185,481–492.https://doi.org/10.1111/j.1469‐8137.2009.03092.x

RCoreTeam.(2018).R: A language and environment for statistical comput‐ing.Vienna,Austria:RFoundationforStatisticalComputing.

Resende,A.F.,Schöngart,J.,Streher,A.S.,Ferreira‐Ferreira,J.,Piedade,M.T.F.,&Silva,T.S.F.(2019).Massivetreemortalityfromfloodpulsedis-turbancesinAmazonianfloodplainforests:Thecollateraleffectsofhy-dropowerproduction.Science of the Total Environment,659,587–598.https://doi.org/10.1016/j.scitotenv.2018.12.208

Richter,B.D.,Baumgartner,J.V.,Powell,J.,&Braun,D.P.(1996).Amethodforassessinghydrologicalterationwithinecosystems.Conservation Biology,10,1163–1174.https://doi.org/10.1046/j.1523‐1739.1996.10041163.x

Ryser,P.,Gill,H.K.,&Byrne,C.J.(2011).ConstraintsofrootresponsetowaterlogginginAlisma triviale. Plant and Soil,343,247–260.https://doi.org/10.1007/s11104‐011‐0715‐0

Scholz, A., Klepsch,M., Karimi, Z., & Jansen, S. (2013).How to quan-tifyconduitsinwood?Frontiers in Plant Science,4,1–11.https://doi.org/10.3389/fpls.2013.00056

     |  2651Journal of Applied EcologyMOSER Et al.

Sheil,D.,Burslem,D.F.R.P.,&Alder,D.(1995).Theinterpretationandmisinterpretationofmortalityratemeasures.Journal of Ecology,83,331–333.https://doi.org/10.2307/2261571

Silva,A.C.D.,Berg,E.V.D.,Higuchi,P.,&Nunes,M.H.(2011).Treecom-munitydynamicafterfloodinginforestfragmentsinsouthernMinasGerais,Brazil.Revista Árvore,35,883–893.https://doi.org/10.1590/S0100‐67622011000500014

Smith,D.M.(1954).Maximummoisturecontentmethodfordeterminingspecificgravityofsmallwoodsamples.ForestProductsLaboratoryreport.Madison,No.2014.

Taiz,L.,&Zeiger,E. (2006).Plant physiology (4thed.).Sunderland,MA:SinauerAssociates.

Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel, C.,Hummel, I., & Garnier, E. (2007). Let the concept of trait befunctional! Oikos, 116, 882–892. https://doi.org/10.1111/ j.2007.0030‐1299.15559.x

Williamson,G.B.,Laurance,W.F.,Oliveira,A.A.,Delamônica,P.,Gascon,C.,Lovejoy,T.E.,&Pohl,L.(2000).Amazoniantreemortalityduringthe1997ElNinodrought.Conservation Biology,14,1538–1542.https://doi.org/10.1046/j.1523‐1739.2000.99298.x

Wittmann, F., Schongart, J., Montero, J. C., Motzer, T., Junk, W. J.,Piedade, M. T. F., … Worbes, M. (2006). Tree species composi-tion and diversity gradients in white‐water forests across theAmazonBasin.Journal of Biogeography,33,1334–1347.https://doi.org/10.1111/j.1365‐2699.2006.01495.x

Wright, I. J., Reich, P. B.,Westoby,M., Ackerly,D.D., Baruch, Z.,Bongers, F.,…Villar, R. (2004). Theworldwide leaf economicsspectrum.Nature,428,821–827.https://doi.org/10.1038/nature02403

Zanne, A. E.,Westoby,M., Falster, D. S., Ackerly, D. D., Loarie, S. R.,Arnold,S.E.J.,&Coomes,D.A.(2010).Angiospermwoodstructure:Globalpatternsinvesselanatomyandtheirrelationtowooddensityandpotentialconductivity.American Journal of Botany,97,207–215.https://doi.org/10.3732/ajb.0900178

SUPPORTING INFORMATION

Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle.

How to cite this article:MoserP,SimonMF,deMedeirosMB,GontijoAB,CostaFRC.Interactionbetweenextremeweathereventsandmega‐damsincreasestreemortalityandaltersfunctionalstatusofAmazonianforests.J Appl Ecol. 2019; 56:2641–2651. https://doi.org/10.1111/1365‐2664.13498