Functional Ecology. 2019;00:1–11. wileyonlinelibrary.com/journal/fec  | 1© 2019 The Authors. Functional Ecology © 2019 British Ecological Society

Received:25March2019  |  Accepted:15July2019DOI: 10.1111/1365-2435.13421

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

Species‐specific effects of phosphorus addition on tropical tree seedling response to elevated CO2

Jennifer B. Thompson1,2 | Martijn Slot1  | James W. Dalling1,3 | Klaus Winter1 | Benjamin L. Turner1  | Paul‐Camilo Zalamea1,4

JenniferB.ThompsonandPaul‐CamiloZalameacontributedequallytothiswork.

1SmithsonianTropicalResearchInstitute,PanamaCity,RepublicofPanama2DepartmentofEnvironmentalScience,Policy,andManagement,UniversityofCaliforniaBerkeley,Berkeley,CA,USA3DepartmentofPlantBiology,UniversityofIllinois,Champaign‐Urbana,IL,USA4DepartmentofIntegrativeBiology,UniversityofSouthFlorida,Tampa,FL,USA

CorrespondencePaul‐CamiloZalameaEmail:camilozalamea@gmail.com

JenniferB.ThompsonEmail:jenniferbriana.thompson@gmail.com

Funding informationSimonsFoundation,Grant/AwardNumber:429440,WTW;NSF‐REU‐SmithsonianTropicalResearchInstitute,Grant/AwardNumber:STRI‐1359299;UniversityofIllinois

HandlingEditor:RebeccaOstertag

Abstract1. Tropicalforestproductivityisoftenthoughttobelimitedbysoilphosphorus(P)availability.Phosphorusavailabilitymightthereforeconstrainpotentialincreasesin growth as the atmospheric CO2 concentration increases, yet there is littleexperimentalevidencewithwhichtoevaluatethishypothesis.WehypothesizedthatwhileallspecieswouldrespondmorestronglytoelevatedCO2whensuppliedwithextraP,theresponsesofindividualspecieswouldalsodependontheirhabi-tatassociationswitheitherhigh‐orlow‐Psoils.WefurtherhypothesizedthatthiseffectwouldbeexacerbatedbyareductionintranspirationrateunderelevatedCO2,astranspirationmayaidinPacquisition.

2. WeusedapotexperimenttotesttheeffectsofPadditiononthephysiologicalandgrowthresponsetoelevatedCO2ofeighttropicaltreespecieswithcontrast-ingdistributionsacrossasoilPgradientinPanamanianlowlandforests.Seedlingsweregrowninanambient(400ppm)orelevated(800ppm)CO2‐controlledglass-houseineitherahigh‐orlow‐PtreatmenttoquantifytheeffectsofPlimitationonrelativegrowthrate(RGR),transpiration,maximumphotosyntheticrateandfoliarnutrients.

3. WefoundevidenceoflimitationbyPandCO2ongrowth,photosynthesis,foliarnutrientsand transpiration.However, theaffinityofaspecies forP,definedasthespeciesdistributionrelativetoPavailability,wasnotcorrelatedwithRGRortranspirationresponsestoelevatedCO2ineitherthelow‐Porhigh‐Ptreatments.

4. Transpiration ratesdecreasedunder elevatedCO2, but foliarPwas greater forsomespeciesunderelevatedCO2,suggestingagreatercapacityforupregulationofPacquisitioninspeciesassociatedwithlow‐Psoils.

5. OurresultsshowthattropicalforestresponsestoelevatedCO2willbespecies‐specificandnotnecessarilyexplainedbyPaffinitiesbasedondistribution,whichposeschallengesforpredictionsofcommunity‐wideresponses.

K E Y W O R D S

climateresponse,CO2fertilization,phosphoruslimitation,speciesdistributions,tropicalforest

2  |    Functional Ecology THOMPSON eT al.

1  | INTRODUC TION

Tropicalforestsstoreoverhalfoftheworld'sforestcarbonstocks(Panetal.,2011)andhavesomeofthehighestratesofnetprimaryproductivityofanybiome(Crameretal.,1999).Thus,tropicalforestshavethepotential tostrongly impactatmosphericCO2concentra-tionsthroughcarbonsequestration.Forexample,withoutaccount-ingforemissionsduetolandconversion,itisestimatedthattropicalforestssequester1.4±0.4PgCyear−1(Schimel,Stephens,&Fisher,2015),whichisequivalentto15%ofglobalannualanthropogenicCemissions.Nonetheless,theresponsesoftropicaltreestoelevatedCO2areunderstudiedcomparedtotemperateecosystems(Cernusaketal.,2013;Leakey,Bishop,&Ainsworth,2012)and,consequently,predictionsofsequestrationpotentialfortropicalforestsunderel-evatedCO2varywidely(Coxetal.,2004;Hickleretal.,2008;Yang,Thornton,Ricciuto,&Hoffman,2016).Thishighlightsuncertaintiesregardingthefutureoftropicalforests,aswellastheirpotentialtomitigatecurrentincreasesinatmosphericCO2.

Itisexpectedthatby2100,atmosphericCO2concentrationswillhave increased from the currentc. 400ppm tobetween538and936ppm,accordingtotheIPCC'sRCP4.5andRCP8.5,respectively(IPCC,2013).ManyplantspecieshavehigherphotosyntheticratesunderelevatedCO2,becausephotosynthesis is lesslimitedbycar-bonavailability(Ainsworth&Rogers,2007;Norbyetal.,2005;Ziska,Hogan,Smith,&Drake,1991).ThismaycauseaCO2fertilizationef-fect,wherebythestimulatedphotosyntheticrateleadstoincreasedgrowth.However,whenonephysiologicallimitationislifted,itisex-pected thatother environmental factorswill limit plant responses(Leakeyetal.,2012),anideaheldbyLiebig'slawoftheminimuminwhichgrowthiscontrollednotbythetotalresourcesavailablebutratherbythelimitingresource.Intemperateecosystemswheresoilnitrogen(N)isoftenthemostlimitingnutrient(Vitousek&Howarth,1991),lowsoilNavailabilitylimitsplantgrowthfromCO2fertiliza-tionovertimeasNbecomesdepletedbystimulatedgrowth(Norby&Warren,2010;Orenetal.,2001;Reichetal.,2006).Incontrast,theproductivityof lowlandtropicalforests isgenerallythoughttobelimitedbysoilphosphorus(P),ratherthanbyN(Clevelandetal.,2011; Turner, Brenes‐Arguedas, & Condit, 2018; Vitousek, 1984).Although N can limit growth in regenerating forests in easternAmazonia (Davidsonetal.,2004)and there is increasingevidenceof nutrient colimitation in the tropics (Wright, 2019), the stronglyweathered nature of the tropical landscapemeans that P ismorelikelytolimitgrowth(Vitousek,2004).Indeed,recentevidencesug-geststhatsuchlimitationoccursbroadlyatthespecieslevel(Turneretal.,2018).

Phosphorus limitationhasstrongphysiologicalandecologicaleffectsonvegetation.Quesadaetal.(2012)foundthatwoodpro-ductionintheAmazonwaspositivelycorrelatedwithtotalsoilP,whileinPanamathemajorityoftreespeciesgrowfasterwheresoilPconcentrationsaregreater(Turneretal.,2018).Phosphorushasalsobeenshowntobeimportantinexplainingtreespeciesdistri-butions(Condit,Engelbrecht,Pino,Perez,&Turner,2013;Pradaetal.,2017;Turneretal.,2018;Zalameaetal.,2016).InthePanama

Canalwatershed,wheresmall‐scaleheterogeneity insoilPavail-ability is comparable to its variation across the entire Amazonbasin,Pavailabilitywasthemost importantedaphicpredictorofthe distribution of over 500 tree species even when comparingothersoilnutrients includingNandK(Conditetal.,2013).Morethanhalfofthespeciesshowedstrongpositiveornegativeasso-ciationswithPavailability (Conditet al.,2013).While increasingPappearstostimulategrowthratesformosttreespeciesinlow-landforestsinPanama(Turneretal.,2018;Zalameaetal.,2016),low‐Plevelsdonotnecessarilyimplyacommunity‐widelimitationonproductivity;rather,somespeciesthriveon low‐Psoils,whileothersdominatesoilswithhigherPlevels(Turneretal.,2018).Infact,Zalameaetal.(2016)showedthatthegrowthofspeciesnat-urally distributed in siteswith high‐P availabilitywas stimulatedbyPadditioninapotexperiment,whilespeciesfromlow‐Pavail-abilitysiteshavelittleornochangeingrowthwhenPwasadded.Furthermore,speciesdistributionalassociationswithsoilPdidnotpredictbiomassallocationorfoliarPwhenplantsweregrownateitherlow‐orhigh‐P(Zalameaetal.,2016).Althoughitisstillun-clearhowsomespeciesmaintainhighgrowthratesonlow‐Psites,this most probably involves mechanisms that promote efficientuse or uptake of P (Turner et al., 2018). Transpiration—the rateofwater lossperunit leafarea—typicallydecreaseswithincreas-ing atmospheric CO2 concentration (e.g.Winter, Aranda, Garcia,Virgio,&Paton,2001).Cernusak,Winter,andTurner(2011)foundthatinresponsetodecreasedtranspirationatelevatedCO2 foliar Pofsometropicaltreespeciesmaydecline.Thus,constraintsbynutrient limitation on the CO2 responsemay be further exacer-batedbynegativeeffectsofelevatedCO2on transpiration rates(Morison&Gifford,1984;Carlson&Bazzaz,1980;Winter,Garcia,Gottsberger,&Popp,2001)giventhattranspirationmayhelpmod-ulate nutrient acquisition, including P, via mass flow (Cernusak,Winter,&Turner,2011).

IncorporatingP limitation intomodelsofplant response to in-creases inCO2 can strongly influencepredictionsof future forestproductivity(Yangetal.,2016).Yangetal.(2016)foundthatinclud-ingthenaturalgradientofsoilPacrosstheAmazonbasinreducedpriorestimatesoffutureforestproductivityresponsestoincreasedCO2byasmuchas26%.Despitethesemodelpredictionsoftheim-portanceofPlimitationinthecontextofglobalclimatechange,fewexperimentalstudieshaveexploredinteractingeffectsofsoilPandCO2ontropicaltrees.InasubtropicaleucalyptusforestinAustralia,Ellsworthetal.(2017)foundthatincreasingtheCO2concentration150ppmaboveambientinaFree‐AirCO2Enrichment(FACE)exper-imentdidnotincreasetreebiomassinP‐limitedsoils.WhentheplotswerefertilizedwithPatambientCO2,however,netprimaryproduc-tivity increased by 35%, suggesting that species response toCO2 wasPlimited.InastudyincentralPanama,twolowlandtreespe-cies,Ficus insipida and Virola surinamensis, accumulated52%morebiomass in open‐top chambers in the presence of twice‐ambientCO2whengivenaNPK+micronutrient fertilizer thanthosegrownwithout fertilizer, indicating nutrient limitation of the response toCO2fertilization(Winter,Garcia,etal.,2001).

     |  3Functional EcologyTHOMPSON eT al.

Hereweconductedagreenhouseexperimentusingambientandtwice‐ambientCO2,and low‐andhigh‐Psoil treatments, todeterminewhether soil P availability limits growth and physio-logical responses to elevated CO2. Because plant responses toPmightvarydependingonadaptationstotheavailabilityofPintheir native habitat (Zalamea et al., 2016), we selected speciesthatarenaturallyassociatedwithhabitatsthatrepresentarangeof soilPavailability (Conditet al., 2013).Wehypothesized thatwhile all specieswould respondmore strongly to elevatedCO2 whensuppliedwithextraP, the responsesof individual specieswouldalsodependontheirhabitatassociationswitheitherhigh‐orlow‐Psoils(Figure1).WepredictedfourpossiblescenariosforplantresponsestoCO2 inrelationtosoilPavailability, inwhichspecies' growth and physiological responses reflect differentialallocation to soil P acquisition. First, a CO2 responsewould beprecluded if specieswithdistributional affinities for low‐PsoilshaveconservativeresourceinvestmentinPuptakeunderlimitedPavailability. Incontrast, specieswithaffinities forhigh‐PsoilsandhigherPneedsmightbeunabletorespondtoelevatedCO2 onlow‐PsoilsduetoP limitation, leadingto littlechangeinthemagnitudeoftheCO2responsewithPaffinity(Figure1,scenario1).Second,theinherentabilityofspeciesassociatedwithlow‐Psoilstobeproductiveatlow‐PavailabilitycouldpermitastrongpositiveCO2response,leadingtoadeclineintheresponsetoCO2 asPaffinity increases (Figure1, scenario2).Third,amorecon-servativegrowthstrategy in speciesassociatedwith low‐PmayconstraintheresponsetoelevatedCO2relativetospeciesassoci-atedwithhigh‐Psoils,leadingtoanincreasingresponsetoCO2asPaffinityincreases(Figure1,scenario3).Finally,intheabsenceof P limitation, theCO2 responsemaybe independent of traitsassociatedwithPuptakeanduse,leadingtolittlevariationintheincreasingresponsetoCO2withPaffinity(Figure1,scenario4).Furthermore,wepredictedthatelevatedCO2wouldreducetran-spirationratesperunitofplantbiomass,resultinginareductioninfoliarPconcentration.ForspeciesassociatedwithPrichsoils,thispredictedreductioninfoliarPwouldfurtherconstraintheirCO2responseunderlow‐Pconditions.

2  | MATERIAL S AND METHODS

2.1 | Study species and growth conditions

SeedsofeightpioneertreespecieswerecollectedfromtheBarroColoradoNatureMonument (BCNM) in central Panama (Table 1).We selected pioneer species because they are ubiquitous in theneotropics, anddue to their small seed reserves, they rapidly be-comedependentontheacquisitionofexternalnutrients.Assuch,theyareidealforstudyingseedlingnutrientdynamics.Furthermore,pioneersthatoccurintheBCNMandacrossthePanamaCanalwa-tershedforestshavewidelyrangingsoilPassociationsthatpredicttheirgrowthresponsetoPadditionunderambientCO2conditions(Zalamea et al., 2016). Using presence/absence data for 550 treespecies across the Panama Canal watershed, Condit et al. (2013)implementedGaussianlogisticregressionmodelstodeterminespe-cies'distributionalassociationswithplant‐availableP.Theyshowedthat resin‐extractable phosphorus, ameasure of plant‐available P,wasastrongpredictorofspeciesdistributions,andmorethanhalfofthetreespeciesstudiedhadpronouncedassociationswitheitherhigh‐orlow‐Psoils.PhosphorusassociationsforeachspecieshavebeenmeasuredasPeffectsizes,definedasthefirst‐orderparam-eterofthe logisticmodelforthespeciesdistributionrelativetoP,whichreflectthechangeintheprobabilityofoccurrenceofaspe-ciesacrossthegradientinPavailabilityinPanamaCanalwatershedforestswhenotherresourcesareheldconstant(Conditetal.,2013).Plant‐availablePvariesfrom<0.1to>20mgP/kgacrossthesefor-ests(Conditetal.,2013).Effectsizevalues<–0.6and>0.6indicatespeciesnaturallyfoundonsoilswith lowandhighavailabilityofP,respectively,valuesbetween–0.6and0.6arereferredtoas“weak”associations.Effect sizes for species included in this study rangedfrom–1.08to1.18(Table1).

Seedsweregerminatedinacommercialpottingsoilinanopen‐air shade house. After c. 3 weeks, seedlings were transplantedinto30cmtall,2.65Lblacktreepots(StueweandSonsInc.)filledwith a 50:50 soil:sand mix. Soil was collected from the SantaRitaRidge,Panama,asitewithoneofthelowestsoilPvaluesinthecanalwatershed.Mean(±SE)extractableresinPforthesitewas0.16±0.03mg/kg,andtotalsoilPwas128±10mg/kg (B.Turner, unpublished data). Seedlingswere grown in glasshousesin the Santa Cruz plant research facility in Gamboa, Panama.Split air‐conditioning units regulated glasshouse temperaturestomatchoutdoor temperatures (c. 32°Cdaytime, c. 24°Cnight‐time), andCO2 concentrationsweremaintainedateither400or800ppm.CO2concentrationwasmaintainedat800ppmbyusingaGMW21Dcarbondioxide transmitter (Vaisala) and aCR‐5000measurementandcontrol system (Campbell Scientific). InjectionofpureCO2wasinitiatedwhentheCO2concentrationfellbelow790 ppm andwas terminatedwhen CO2 concentration reached800ppm.TheCO2concentrationwaspreventedfromovershoot-ingbyprovidingCO2 inmultiplepulsesof2s interruptedby5swithoutCO2injection.Seedlingswerewatereddailyandreceived150mlofaliquidfertilizationtreatmentweekly(asinZalameaet

F I G U R E 1  PotentialresponsestoelevatedCO2amongspecieswithcontrastingdistributionalaffinitiesforsoilPindifferentsoilPenvironments.ThemagnitudeofCO2responseindicatesthedegreeofchangebetweentheambientandelevatedCO2treatments(seetextfordetails)

4  |    Functional Ecology THOMPSON eT al.

al.,2016).Allseedlingsreceivedafullnutrienttreatment includ-ing4mMKNO3,1.5mMMgSO4and4mMCaCl2,micronutrientsand Fe as ethylenediaminetetraacetic acid iron (III) sodium salt.Inaddition,plantsgrown in the+P treatment received1.33mMNaH2PO4perweek,whilethoseinthe–Ptreatmentdidnot.Therewere six replicate seedlingsper species foreachcombinationofCO2 andP treatments.Theseedlingsweremoved toadifferentlocation in the chamber at least twiceduring the experiment toreducepotentialeffectsofbenchlocation.

2.2 | Relative growth rate measurements

Before fertilizing the seedlings for the first time, five individualsperspecieswereharvestedanddriedinanovenat70°Cfor3days.These initialweightswereused tocalculatemean relativegrowthrate(RGR)overtheexperiment.

RGR=

ln (finalmass)− ln (initialmass)Days of growth

Seedlingsgrewfor27–62daysdependingonthespeciesgrowthrate (Table 1). We aimed for plant biomass to pot volume ratiosto stay, on average, <1 g/L in the+P treatment, following recom-mendations of Poorter, Bühler, Dusschoten, Climent, and Postma(2012).Leafareawasmeasuredwithanautomatedleafareameter(LI‐3000A;LI‐COR).Seedlingsweredriedinanovenat70°Cforatleast3daysbeforebeingweighed.

2.3 | Whole‐plant transpiration rates

Wemeasuredwhole‐planttranspirationratesgravimetrically2–3daysbeforeharvest.Plantswerewatereduntilsoilsaturationandpotssealedwithplasticbagsaroundtheseedlingstembasetopreventsoilwaterevaporation.Potswereweigheddirectlyafterwateringand48hrlatertodeterminetotaldailywaterlossperunitleafarea(gcm–2 24 hr–1).

2.4 | Foliar nutrient analysis

Leaveswere dried, ground and analysed for total foliarN and P.TotalfoliarPwasdeterminedbyfirstashingleaftissueinamuffle

furnace (550°C for 1 hr), and then dissolving in 1MHCl,with PdetectionbymolybdovanadatecolorimetryonaLachatQuikchem8500(HachLtd).TotalfoliarNwasmeasuredbydrycombustiononaThermoFlashEA1112analyzer(CEElantech).

2.5 | Photosynthetic capacity

Netphotosynthesiswasmeasuredatapre‐determinedsaturatingirradiancelevelof1,200µmolm–2s–1attheCO2concentrationsoftheglasshouse,thatis,bothat400and800ppm.Measurementswere made with a LI‐6400XT portable photosynthesis system(LI‐COR)ononefullyexpanded, recentlymatured leafperplantat30.3±0.3°C(mean±SDofleaftemperatureduringmeasure-ment).Leaf‐to‐airvapourpressurewasmaintainedbelow2.5kPaduring all measurements. Not all treatment combinations pro-duced leaves large enough to be enclosed in the 2 × 3 cm leafcuvette.Intotal,datawereobtainedfor140plantsbelongingto8species,66inthe–Ptreatmentand74inthe+Ptreatment.

2.6 | Statistical analysis

Weusedlinearregressionanalysestoestimatetherelationshipbetween the P effect size and the per cent increase in RGR,maximumphotosyntheticrate,foliarnutrientsandtranspirationin response toP fertilizationandCO2 treatments.Regressionswereperformedonthepercent increase intheresponsevari-able to fertilization at 800 relative to 400 ppm. The per centincreasewas calculated using the average values from each Ptreatmentas:

ThepercentincreaseinresponsetoelevatedCO2inthe+Prel-ativetothe–Ptreatmentwasalsocalculated,resultsofwhicharereportedintheSupportingInformation(AppendixS1).

WeadditionallyusedlinearmixedeffectmodelstotestwhetherPfertilization,CO2levelandtheCO2byPinteractionaffectedthedif-ferentvariables:RGR,foliarnutrients,transpirationrateandmaximumphotosyntheticrate.PhosphorusandCO2treatmentsweresetasfixed

% increase=800−400

400×100.

Species FamilyGrowth period (days) P effect size

Trichospermum galeottii(Turcz.)Kosterm.

Malvaceae 62 –1.08

Cecropia insignisLiebm. Urticaceae 55 –0.86

Ochroma pyramidale(Cav.exLam.)Urb. Malvaceae 44 –0.15

Trema micrantha(L.)Blume Cannabaceae 27 0.37

Ficus insipidaWilld. Moraceae 62 0.48

Guazuma ulmifoliaLam. Malvaceae 41 0.69

Cecropia peltataL. Urticaceae 53 0.99

Cecropia longipesPittier Urticaceae 51 1.18

Note: MorenegativePeffectsizesindicatedistributionalassociationswithsoilswithlowerPavail-ability(Conditetal.,2013).

TA B L E 1  Speciesusedinthegreenhouseexperimentandtheirrespectivefamily,growthperiodandPeffectsizes

     |  5Functional EcologyTHOMPSON eT al.

effectsandspeciesasrandom.Allnon‐normallydistributeddatawerelogtransformedbeforeanalysis.Datawereanalysedwiththepackagenlmeandthelme functioninR(version3.5.10).

3  | RESULTS

3.1 | Overall treatment effects

TheCO2andPtreatmentshadsimilarlystrongpositiveoverallef-fects on seedling growth (RGR) across species (Figure2, Table2).However,therewasnoevidencethatthePtreatmentimpactedplantresponsetoCO2(i.e.nosignificantCO2byPtreatmentinteraction;Figure2,Table2).Similarresultswerefoundforwhole‐planttran-spirationandfoliarNconcentration,withreducedratesandconcen-trationsunder elevatedCO2 andP treatments (Figure2, Table2).Despite reductions in foliar N, the maximum photosynthetic rate(Amax)wasnearlydoubledunderelevatedCO2,butAmaxdidnotre-spondtothePtreatmentnorshowaCO2byPinteraction(Figure2,Table2).FoliarPconcentrationsrespondedpositivelytothePandCO2treatmentsandshowedastrongtreatmentinteraction;asimilarpatternwasobservedforfoliarN:P,whichrangedfrom>30intheambientCO2andlow‐Ptreatmentto<15inthehighCO2andhigh‐Ptreatments(Figure2,Table2).FoliarC:NincreasedwithPadditionandelevatedCO2(Table2).

3.2 | Species‐level responses

ResponsestoPandCO2treatmentsvariedatthespecieslevelandwereoftenindependentofthedistributionalassociationwithsoilP(Figure3).Wefoundstrongsupportforscenarios1and4(Figure1),indicating that species differences in allocation toP uptake andPlimitationunderelevatedCO2resultedofteninanabsenceofinter-active effects between fertility and the response toCO2.Despitelarge differences in growth responses among species, we did notfindthatthesoilPassociationofspeciespredicteditsRGRresponsetoelevatedCO2eitherinthelow‐Ptreatment(R

2=0.26,p=0.11,n=8;Figure3a)orinthehigh‐Ptreatment(R2=−0.11,p=0.59,n=8;Figure3a).Despitetheabsenceofapredictablegrowthresponse, in-dividual traitsdid reflectspeciesdistributions.Therewasasignifi-cantnegativecorrelationbetweenspeciesPaffinityandthepercentincreaseinmaximumphotosyntheticratefromambienttoelevatedCO2 for the +P treatment (R

2 = 0.55, p = 0.03, n = 7; Figure 3b),butnotforthe–Ptreatment(R2=0.23,p=0.19,n=6;Figure3b).Therefore,thephotosyntheticrateofspeciesassociatedwithlow‐PsoilswasmorestronglystimulatedbyelevatedCO2inthepresenceofextraPcomparedtospeciesassociatedwithhigh‐Psoils.Whilepho-tosyntheticratesmeasuredattheirtreatmentCO2levelwerealwaysgreaterfortheelevatedCO2plants,whenphotosynthesiswascom-paredat400ppmCO2,theratestendedtobelowerforplantsgrown

F I G U R E 2  AveragetraitvaluesforspeciesacrossCO2andPtreatments:(a)relativegrowthrate(RGR),(b)maximumphotosyntheticrate,(c)transpiration,(d)totalfoliarP,(e)totalfoliarNand(f)foliarN:Pratio.Errorbarsrepresent±1SE.Foreachtrait,statisticalsignificanceofeachtreatment(i.e.CO2,PandCO2byPinteraction)isnotedwithasterisks(seeTable2fordetails)

6  |    Functional Ecology THOMPSON eT al.

atelevatedCO2.Thiseffectwasparticularlystronginthe–Ptreat-ment(datanotshown),indicatingthatphotosyntheticcapacitywasdownregulatedathighCO2supply,especiallyunderlimitingnutrientconditions.TherewasnorelationshipbetweenspeciesPassociationandincreaseintranspirationrateinthe–P(R2=−0.04,p=0.44,n=8;Figure3c),nor+P(R2=−0.16,p=0.87,n=8;Figure3c)treatments.

FoliarPfollowedasimilar trendtophotosynthesis. Inthe+Ptreatment, there was a marginally significant negative relation-shipbetweenPaffinityand theproportional increase in foliarPconcentrationbetweenthe400and800ppmchamber(R2=0.40,p = 0.05, n = 8; Figure 3d). Species associatedwith low‐P soilstherefore increased foliar P more when fertilized under ele-vated CO2 than species associatedwith high‐P soils. Therewasno relationship in the–P treatment (R2=−0.04,p=0.43,n = 8;Figure3d).AsimilarpatternwasobservedforfoliarN,withaneg-ative relationshipwithPaffinity in the+P treatment (R2=0.56,p=0.02,n=8;Figure3e),andnorelationshipinthe–Ptreatment(R2=−0.12,p=0.65,n=8;Figure3e).Finally, therewasamar-ginallysignificant,negativerelationshipbetweentheproportionalincreaseinthefoliarN:PratioandPaffinityinthe–Ptreatment(R2=0.29,p=0.09,n=8;Figure3f),butnosignificantdifferenceinthe+Ptreatment(R2=0.04,p=0.30,n=8;Figure3f).

4  | DISCUSSION

Seedlinggrowth increased in response toPadditionandelevatedCO2, suggesting that growthonP‐poor soils canbe co‐limitedbyPandCO2.Likewise,photosynthesisrateswerehigheratelevatedCO2 and tended tobe stimulatedbyPaddition.These results areconsistentwithpreviousstudiesof responsesofgrowthandpho-tosynthesistoPadditionandCO2fertilizationintropicaltrees(e.g.Cernusak,Winter,Martinez,etal.,2011).However,wedidnotfinda significant interaction between the P and CO2 treatments. Thisresult indicatesthatPadditiondoesnotstimulatetheresponseofplantgrowthandphotosynthesis toCO2 fertilization in thisgroupof tropical pioneer species. This result adds empirical evidence totheresultsoftheCLM4‐CNPmodelofYangetal.(2016)thatfoundthataCO2fertilizationeffectintheAmazonregionwouldbegreatlyoverestimated if soil P availabilitywasnot considered.Themodelconsidered C, N and P cycling aswell as other site‐specific char-acteristics,butemployedplantfunctional types (PFTs)ratherthan species‐specificdata.

The CLM4‐CNP model uses two tropical PFTs, broadleaf ev-ergreen and broadleaf deciduous, which are separated by pho-tosynthetic parameters, stem and leaf optical properties and rootdistributionpatterns,amongothers.Thus,themodelassumesthatalltreespeciesfallintooneofthesetwocategoriesandbehavesim-ilarly intermsofphotosynthesis,resourcesuseandgrowth.Thesebroad categories can overly simplify forests response to elevatedCO2, or not portray a completely accurate one, aswe found thatevenwithinonePFT(here,earlysuccessional,mostlybroadleafev-ergreen species) vary greatly in their responses to environmentalchanges.Belowwedescribe thepotential forP limitation todrivespecies‐specificresponsestorisingatmosphericCO2.

4.1 | How does soil P availability modulate seedling response to elevated CO2?

Overall,elevatedCO2decreasedtranspirationrates,consistentwithmanystudies(e.g.Morison&Gifford,1984;Carlson&Bazzaz,1980;Winter,Aranda,etal.,2001).WealsosawaneffectofPtreatmentontranspiration,withreducedratesinthe+Ptreatment.Cernusak,Winter, and Turner (2011) found a positive correlation betweenthe transpiration ratio (mass ofwater transpired per plant carbonmassproduced)andPcontentintropicaltreeseedlingsinPanama,suggesting that higher transpiration rates increase P acquisitionviamassflow.Cernusak,Winter,andTurner(2011)alsofoundthatwhentranspirationwasreducedbyelevatedCO2,theleafC:PratioincreasedinSwietenia macrophylla,butstayedconstantonOrmosia macrocalyx.ThissuggeststhatfoliarPconcentrationsofsometropi-caltreespeciesmaydeclineinresponsetodecreasingtranspirationrates,whereas other speciesmay be unaffected.Our results pro-videonlypartial support for this; in the–P treatment, foliarPdidnotdifferbetweenCO2treatments,whileinthe+Ptreatment,foliarPwasfurtherincreasedatelevatedCO2.Thesecontrastingresultshighlight thecomplexityofPuptakemechanisms,withapotential

TA B L E 2  Resultsoflinearmixed‐effectsmodelsforeachtrait

Trait

Factor

CO2 treatment P treatment CO2 ∙ P treatment

Relativegrowthrate

F 47.95 40.15 0.09

p <0.0001 <0.0001 0.76

Photosynthesis

F 112.89 2.82 0.93

p <0.0001 0.09 0.34

Transpiration

F 51.12 21.59 2.19

p <0.0001 <0.0001 0.14

FoliarP

F 14.10 80.66 14.10

p <0.001 <0.0001 <0.001

FoliarN

F 22.83 8.12 0.15

p <0.0001 <0.01 0.69

N:P

F 31.80 92.51 4.57

p <0.0001 <0.0001 0.03

C:N

F 18.00 5.333 0.0091

p <0.0001 0.022 0.9240

Statisticalsignificanceofeachtreatment(i.e.CO2,P,andCO2byPinteraction)isnotedinbold.

     |  7Functional EcologyTHOMPSON eT al.

roleformassflowinsomespecies,butnotothers.Furthermore,thespecies‐specificnatureofCO2effectsonPuptakecouldhavemajorconsequencesforcompetitiveinteractionsamongspeciesunderdif-ferentPenvironmentsandclimatechangescenarios.

Nutrient addition experiments in the field (Ostertag, 2010;Santiago et al., 2011;Wright et al., 2011) and in pot experiments(Zalameaetal.,2016)haveshownthatfoliarPincreaseswithPaddi-tionand,asexpected,foliarPconcentrationincreasedwithPaddi-tioninthecurrentstudy.FoliarPalsoincreasedunderelevatedCO2. Moreover,PandCO2treatmentsshowedastronginteractioneffect;onlywhenPwassupplieddidfoliarPincreaseunderelevatedCO2. Tang,Chen,andChen(2006)foundanincreaseinfoliarPoftropicalweedyforbspecieswithelevatedCO2,whichtheyattributedtoanincreaseinmycorrhizalcolonizationandPuptakeasplantshadin-creasednutrientdemandfromthestimulatedgrowth.Carbohydratesupply to symbionts becomes less costly for plants when grownathighCO2, leadingtohigherfoliarPunderelevatedCO2butnotunderambientCO2.Ourplantsweregrownonunsterilizedsoilcol-lectedfromthefield,soitislikelythatourplantswerecolonizedbymycorrhizalfungi.Nasto,Winter,Turner,andCleveland(2019)alsofoundincreasedAMFcolonizationandphosphataseactivityunderelevatedCO2, whichwas attributed to the trees having a greatermetaboliccapacitytoacquireP.Inourpotexperiment,wedidnotfindarootmassfractionresponsetoelevatedCO2(datanotshown),althoughLiuetal.(2013)foundthatinasubtropicalforestinChina,

increasedrootgrowthunderelevatedCO2 increasedPacquisitionand foliar P. Speciesmay overcome low soil P availability by spe-cializingintheacquisitionofdifferentformsofP,includingorganicPforms (Nastoetal.,2017;Steidinger,Turner,Corrales,&Dalling,2015),whichcouldunderliedifferences inphysiological responsestoelevatedCO2.

FoliarNandN:PratioswerereducedunderelevatedCO2andPaddition.Phosphorusadditionsignificantlyincreasedgrowthrates,andwithout additionalN supply, the increase in growth ratemayhaveledtoNdilutioninthetissueandanincreaseintheC:Nratio.Areduction in foliarN iscommonlyobservedunderelevatedCO2 andmay be associatedwith a reduction ofN‐rich photosyntheticenzymesorincreasedstarchcontentofleaves(Medlynetal.,2002).When photosynthetic rates weremeasured on plants at their re-spective CO2 treatment concentrations, rates were higher underelevatedCO2.However,whencomparedatthesameCO2 concen-tration, plants grown at elevated CO2 had lower photosynthesisrates,suggestingadown‐regulationofphotosyntheticcapacity.

Photosynthetic rateswereonlymarginally affectedbyP addi-tion in our experiment; however, we found a positive correlationbetweenfoliarPandphotosyntheticrates(SeeAppendixS2),indi-catingthatinadditiontoCO2limitation,photosynthesiscanalsobePlimited,asshownpreviouslyforplantsinP‐limitedsoils(Lovelocket al., 1997; Raaimakers, Boot, Dijkstra, Pot, & Pons, 1995). Thisis consistent with the view that photosynthesis in tropical trees

F I G U R E 3  RelationshipsbetweenPeffectsizeandthepercentincreasebetweenthe400and800ppmtreatmentsforplantresponses:(a)relativegrowthrate(RGR),(b)maximumphotosyntheticrate,(c)transpiration,(d)totalfoliarP,(e)totalfoliarNand(f)foliarN:Pratio.Significantrelationshipsarenotedwithasolidlineandmarginallysignificantrelationships(p <0.1)withadottedline.Shadedgreybandsrepresent95%CI

0

25

50

−1.0 −0.5 0.0 0.5 1.0

Incr

ease

in R

GR

(%)

(a)

−100

0

100

200

−1.0 −0.5 0.0 0.5 1.0

Incr

ease

in A

max

(%)

(b)

−100

−50

0

50

−1.0 −0.5 0.0 0.5 1.0

Incr

ease

in tr

ansp

iratio

n (%

)

(c)

0

100

200

−1.0 −0.5 0.0 0.5 1.0

Incr

ease

in fo

liar P

(%)

(d)

−50

−25

0

25

−1.0 −0.5 0.0 0.5 1.0P effect size

Incr

ease

in fo

liar N

(%)

(e)

−60

−40

−20

0

20

−1.0 −0.5 0.0 0.5 1.0

Incr

ease

in N

:P (%

)

(f)P treatment

−P

+P

8  |    Functional Ecology THOMPSON eT al.

isoftenco‐limitedbyNandP (Norbyetal.,2016).Lovelocketal.(1997)foundthatunderelevatedCO2andinoculationwithAMF,thetropical treeBeilschmiedia pendula photosynthesizedmore than inambientCO2andnon‐mycorrhizalconditions,likelyduetoPlimita-tionofphotosynthesis.

4.2 | Can species distributional P affinities predict species responses to elevated CO2?

Inapreviouspotexperiment,Paffinitiesofspecies—basedontheirdistributionwithin the PanamaCanal area—were a strong predic-torofspeciesgrowthresponsetoPaddition(Zalameaetal.,2016).Thus,specieswithanaffinity forhigh‐Psoils, indicatedbyaposi-tiveassociationalPeffectsize,showeda larger increase inRGRinresponsetoPadditionthanthosewithanaffinityforlow‐Psoils.Incontrast,herewefoundthatPeffectsizescouldnoteasilypredictthemagnitudeof species growth responses to elevatedCO2. ThemagnitudeofgrowthresponsestoPadditionincreasedwithPeffectsize,bothinambientCO2(asinZalameaetal.,2016)andatelevatedCO2 (seeAppendixS1),butforsomespeciesthegrowthresponsewas greater under elevated CO2, while for others it was greaterunderambientCO2.Overall, thegrowthresponsetoPadditionatambientCO2wasonlymarginallysignificantlycorrelatedwiththatatelevatedCO2(Spearman'srankcorrelation:ρ=0.71,p=0.057,n=8).All species responded to elevatedCO2, but the degree of growthstimulation inthetwoPtreatmentswas independentofPaffinity(Figure3a).Ourresultsstronglysupportscenarios1and4(Figure1)formostofthemeasuredtraits.Thus,whiletherearephysiologicallimitationstoCO2responseintheabsenceofP,thesearegenerallynotdependentondistributionalassociationstoP(Figure1—Scenario1),ortheCO2responseis independentofPaffinitywhenPisnotlimiting(Figure1—Scenario4).

Whilemosttrait responsestoelevatedCO2were independentofdistributionalPaffinities,wefoundasignificantnegativecorrela-tionbetweenPaffinityandtheincreaseinmaximumphotosyntheticratefromambienttoelevatedCO2inthe+Ptreatment.Thisresultisdrivenstronglybythehigherpercentincreaseinmaximumphoto-syntheticrateatelevatedCO2ofthetwospecieswiththelowestPeffectsizesinourexperiment—Trichospermum galeottii and Cecropia insignis.ThesetwospeciesalsohadmuchhigherfoliarPatelevatedthan at ambientCO2 in the+P treatment (c. 110% increase, com-pared to an average c. 37% for the remaining species). Curiously,atambientCO2boththefoliarPandphotosyntheticratesofthesespeciesweremuchlowerunder+Pthanunder–P.ThissuggeststhatphotosynthesiswasstimulatedbyfoliarPregardlessofthespecies'distributionalPaffinity(seeAppendixS2).Thesephotosynthesisre-sultsindicatethattheeffectofelevatedCO2oncarbonfixationoftropicaltreeswillbehighlyspeciesspecific,andthatresponsesmaydependonPavailability.

Tree species normally found in low‐P environments tend tohavehigherP‐useefficiency(PUE)thansoilgeneralists(Gleason,Read,Ares,&Metcalfe,2009).WefoundthatintheambientCO2 andPadditiontreatments,speciesnaturallydistributedinhigh‐P

siteshadhigherconcentrationsofPintheirleaveswhencomparedto speciesnaturallydistributed in low‐P sites (seeAppendixS3),butthisrelationshipdisappearedatelevatedCO2.Furthermore,inthe P addition treatment therewas amarginally significant neg-ative relationshipbetweenP affinity and the increase in foliarPconcentrationbetween ambient andelevatedCO2. In a fertiliza-tionexperimentinlowlandtropicalChina,PadditionincreasedallfourfunctionaltypesoffoliarP:structuralP,metabolicP,nucleicacidPandresidualPoftropicaltrees(Moetal.,2019).Themagni-tudeofchangewasspecies‐specific,butspeciescouldreallocateP pools tomeet photosynthetic demands (Mo et al., 2019), sug-gestingthatsomespeciesmaybeabletoovercomePlimitationofcertainresponsesinP‐poorsoils.Theseresultshighlighttheneedfor furtherstudies tounderstand thedifferences inspeciesPUEorPuptakemechanismsthatallowspeciestothriveinlow‐Penvi-ronmentsandspecificallyhowtheymayaffectspeciesresponsestoelevatedCO2.

Phosphorusaffinitiesoflate‐successionalspeciesareasvariableas those of early‐successional species (Condit et al., 2013). Thus,we hypothesize that strategies to acquire and use soil available Pshouldvarysimilarlyamongspeciesinthesetwofunctionalgroups.At the early seedling stage, late‐successional species are unlikelytorespondstronglytoPaddition,regardlessofCO2concentration,becausethelargerseedsoflate‐successionalspeciessupportseed-linggrowthmuchlongerthanthesmallseedsofearly‐successionalspecies (Slot,Palow,&Kitajima,2013). FoliarP concentration sig-nificantly increased with P addition (Santiago et al., 2011), but itseemsunlikelythatthiselevatedleafPwouldsupportastrongCO2 fertilizationeffectonlate‐successionalspeciesgiventheconserva-tivegrowthhabitsofthisfunctionalgroup.PreviousworkinPanamahasshownthatwhileearly‐successionalspeciesrespondstronglytoCO2enrichment, late‐successionalspeciesdonot,evenwhensup-pliedwith amplemineral nutrients (Winter& Lovelock, 1999). Asourstudyfocusedonsmall‐seeded,early‐successionalspecies,firmconclusionsaboutthecommunity‐levelresponsestoclimatechangewill require future studies that include late‐successional species.However,weexpect thatspecies‐specificeffectsofPadditiononthephysiologicalandgrowthresponsetoelevatedCO2foundinthisstudyforearly‐successionalspeciesarelikelytobefoundonotherfunctionalgroups.

Our study provides insight at the complex interactions be-tweensoilPavailabilityandelevatedCO2intropicaltreespecies.ThisisthefirststudydeterminingresponsestonutrientlimitationandelevatedCO2inrelationtoknownPassociationsinthefield.WhilewefoundbothPandCO2 limitationoverall inthecontextof our pot experiment, individual species responses to elevatedCO2andsoilPwere largelyunrelatedtotheir respectiveassoci-ationswithPavailabilityinthefield.Thiscouldhaveimplicationsfor future forestcomposition,and thereforecarbonstorageandresidencytime,asspeciesresponsestoCO2dependnotonlyontheavailabilityofP,butalsospecies‐specificrelationshipstoP.Atleastfortheeightspeciesstudiedhere,theserelationshipscannotbereadilypredictedfromthePassociationofthespecies inthe

     |  9Functional EcologyTHOMPSON eT al.

field.Theseresultsemphasizetheimportanceofconductingcom-plementary in situexperiments in the tropics (Würth,Winter,&Körner,1998),suchastheplannedAmazonFACEprogram,aswellasstudyingspecies‐specific responses inorder toascertainhowhighly productive tropical forests can sequester CO2 under soilnutrientlimitation.

ACKNOWLEDG EMENTS

WethanktheSmithsonianTropicalResearchInstituteforprovidingfacilities, logisticalsupportandpermissiontoconduct theproject.WeespeciallythankC.Sarmiento,A.Bielnicka,M.Garcia,J.Aranda,J.SalasandD.Agudoforassistanceinthelaboratoryandthegreen-house.WealsothankC.SarmientoforthepreparationofFigure1.Wewould like to thank theAssociateEditor, and twoanonymousreviewers for the comments and suggestions, which have greatlyhelpedtoimproveourmanuscript.JBTwassupportedbyNSF‐REUto the Smithsonian Tropical Research Institute (STRI—1359299).PCZwassupportedbyagrantfromtheSimonsFoundationtoSTRI(429440,WTW). PCZ, MS, BLT and KW were supported by theSmithsonianTropicalResearchInstitute.JWDwassupportedbytheUniversityofIllinois.

AUTHORS' CONTRIBUTIONS

J.W.D.,P.‐C.Z.,K.W.andB.L.T. conceivedanddesigned the study.J.B.T.,P.‐C.Z.,M.S., J.W.D.,K.W.andB.L.T. collected thedataandperformedtheexperiment.J.B.T.,P.‐C.Z.andM.S.analysedthedata.J.B.T.,P.‐C.Z.,J.W.D.andM.S.wrotethemanuscriptandalltheau-thorscommentedonit.

DATA AVAIL ABILIT Y S TATEMENT

RawdataareavailablefromtheDryadDigitalRepositoryathttps://doi.org/10.5061/dryad.777461k(Thompsonetal.,2019).

ORCID

Martijn Slot https://orcid.org/0000‐0002‐5558‐1792

Benjamin L. Turner https://orcid.org/0000‐0002‐6585‐0722

Paul‐Camilo Zalamea https://orcid.org/0000‐0002‐0987‐4164

R E FE R E N C E S

Ainsworth,E.A.,&Rogers,A. (2007).Theresponseofphotosynthesisand stomatal conductance to rising [CO2]: Mechanisms and envi-ronmentalinteractions.Plant, Cell and Environment,30(3),258–270.https://doi.org/10.1111/j.1365‐3040.2007.01641.x

Carlson,R.W.,&Bazzaz,F.A.(1980).TheeffectsofelevatedCO2 con-centrationsongrowth,photosynthesis,transpiration,andwateruseefficiencyofplants.InJ.J.Singh&A.Deepak(Eds.), Environmental and climatic impact of coal utilization. Proceedings of a Symposium, Williamsburg, Virginia, USA April 17–19, 1979 (pp. 609–623). NewYork,NY:AcademicPress.

Cernusak,L.A.,Winter,K.,Dalling,J.W.,Holtum,J.A.M.,Jaramillo,C.,Körner, C.,…Wright, S. J. (2013). Tropical forest responses to in-creasingatmosphericCO2:Currentknowledgeandopportunitiesforfutureresearch.Functional Plant Biology,40(6),531–551.https://doi.org/10.1071/FP12309

Cernusak, L. A., Winter, K., Martinez, C., Correa, E., Aranda, J.,Garcia, M., … Turner, B. L. (2011). Responses of legume versusnonlegume tropical tree seedlings to elevated CO2 concentra-tion. Plant Physiology, 157(1), 372–385. https://doi.org/10.1104/pp.111.182436

Cernusak,L.A.,Winter,K.,&Turner,B.L.(2011).Transpirationmodulatesphosphorus acquisition in tropical tree seedlings. Tree Physiology,31(8),878–885.https://doi.org/10.1093/treephys/tpr077

Cleveland,C.C.,Townsend,A.R.,Taylor,P.,Alvarez‐Clare,S.,Bustamante,M.M.C.,Chuyong,G.,…Wieder,W.R.(2011).Relationshipsamongnetprimaryproductivity,nutrientsandclimateintropicalrainforest:Apan‐tropical analysis.Ecology Letters,14(9),939–947.https://doi.org/10.1111/j.1461‐0248.2011.01658.x

Condit,R.,Engelbrecht,B.M.J.,Pino,D.,Perez,R.,&Turner,B.L.(2013).Speciesdistributionsinresponsetoindividualsoilnutrientsandsea-sonaldrought across a communityof tropical trees.Proceedings of the National Academy of Sciences,110(13), 5064–5068. https://doi.org/10.1073/pnas.1218042110

Cox, P.M., Betts, R. A., Collins,M.,Harris, P. P., Huntingford, C., &Jones, C. D. (2004). Amazonian forest dieback under climate‐carbon cycle projections for the 21st century. Theoretical and Applied Climatology, 78(1–3), 137–156. https://doi.org/10.1007/s00704‐004‐0049‐4

Cramer, W., Kicklighter, D. W., Bondeau, A., Moore, B.III,Churkina, G., Nemry, B. … The Participants of the PotsdamNPP Model Intercomparison (1999). Comparing global mod-els of terrestrial net primary productivity (NPP): Overviewand key results. Global Change Biology, 5(S1), 1–15. https://doi.org/10.1046/j.1365‐2486.1999.00009.x

Davidson,E.A.,ReisdeCarvalho,C.J.,Vieira,I.C.G.,Figueiredo,R.D.O.,Moutinho,P.,YokoIshida,F.,…TumaSabá,R. (2004).Nitrogenand phosphorus limitation of biomass growth in a tropical sec-ondary forest. Ecological Applications, 14(4), 150–163. https://doi.org/10.1890/01‐6006

Ellsworth, D. S., Anderson, I. C., Crous, K. Y., Cooke, J., Drake, J. E.,Gherlenda,A.N.,…Reich,P.B.(2017).ElevatedCO2doesnotincreaseeucalyptforestproductivityonalow‐phosphorussoil.Nature Climate Change,7(4),279–282.https://doi.org/10.1038/nclimate3235

Gleason,S.M.,Read,J.,Ares,A.,&Metcalfe,D.J. (2009).Phosphoruseconomicsoftropicalrainforestspeciesandstandsacrosssoilcon-trasts in Queensland, Australia: Understanding the effects of soilspecialization and trait plasticity. Functional Ecology, 23(6), 1157–1166.https://doi.org/10.1111/j.1365‐2435.2009.01575.x

Hickler, T., Smith, B., Prentice, I. C., Mjöfors, K., Miller, P., Arneth,A., & Sykes, M. T. (2008). CO2 fertilization in temperate FACEexperiments not representative of boreal and tropical for-ests. Global Change Biology, 14(7), 1531–1542. https://doi.org/10.1111/j.1365‐2486.2008.01598.x

IPCC(2013).Climatechange2013.InV.Bex,P.M.Midgley,T.F.Stocker,D.Qin,G.‐K.Plattner,M.Tignor,S.K.Allen,J.Boschung,A.Nauels,& Y. Xia (Ed.), Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.Cambridge,UKandNewYork,NY:CambridgeUniversityPress.

Leakey,A.D.B.,Bishop,K.A.,&Ainsworth,E.A.(2012).Amulti‐biomegapinunderstandingofcropandecosystemresponsestoelevatedCO2. Current Opinion in Plant Biology, 15(3), 228–236. https://doi.org/10.1016/j.pbi.2012.01.009

Liu,J.,Huang,W.,Zhou,G.,Zhang,D.,Liu,S.,&Li,Y. (2013).Nitrogentophosphorusratiosoftreespeciesinresponsetoelevatedcarbon

10  |    Functional Ecology THOMPSON eT al.

dioxideandnitrogenaddition in subtropical forests.Global Change Biology,19(1),208–216.https://doi.org/10.1111/gcb.12022

Lovelock, C. E., Kyllo, D., Popp,M., Isopp, H., Virgo, A., &Winter, K.(1997). Symbiotic vesicular‐arbuscularmycorrhizae influencemaxi-mumratesofphotosynthesisintropicaltreeseedlingsgrownunderelevatedCO2. Australian Journal of Plant Physiology,24(2),185–194.https://doi.org/10.1071/PP96089

Medlyn, B. E., Dreyer, E., Ellsworth, D., Forstreuter,M., Harley, P. C.,Kirschbaum,M.U.F.,…Loustau,D. (2002).Temperatureresponseofparametersofabiochemicallybasedmodelofphotosynthesis.II.Areviewofexperimentaldata.Plant, Cell & Environment,25(9),1167–1179.https://doi.org/10.1046/j.1365‐3040.2002.00891.x

Mo,Q., Li, Z., Sayer, E. J., Lambers,H., Li, Y., Zou, B. I.,…Wang, F.(2019). Foliar phosphorus fractions reveal how tropical plantsmaintain photosynthetic rates despite low soil phosphorusavailability. Functional Ecology, 33(3), 503–513. https://doi.org/10.1111/1365‐2435.13252

Morison,J.I.L.,&Gifford,R.M.(1984).PlantgrowthandwaterusewithlimitedwatersupplyinhighCO2concentrations.I.Leafarea,wateruse and transpiration. Australian Journal of Plant Physiology, 11(5),361–374.https://doi.org/10.1071/PP9840361

Nasto,M. K.,Osborne, B. B., Lekberg, Y., Asner, G. P., Balzotti, C. S.,Porder, S., … Cleveland, C. C. (2017). Nutrient acquisition, soilphosphoruspartitioningandcompetitionamongtrees ina lowlandtropicalrainforest.New Phytologist,214(4),1506–1517.https://doi.org/10.1111/nph.14494

Nasto,M.K.,Winter,K.,Turner,B.L.,&Cleveland,C.C.(2019).NutrientacquisitionstrategiesaugmentgrowthintropicalN2‐fixingtreesinnutrient‐poorsoilandunderelevatedCO2.Ecology,100(4),e02646.https://doi.org/10.1002/ecy.2646

Norby, R. J.,DeLucia, E.H.,Gielen, B., Calfapietra,C.,Giardina,C. P.,King, J. S., …Oren, R. (2005). Forest response to elevatedCO2 isconservedacross abroad rangeofproductivity.Proceedings of the National Academy of Sciences, 102(50), 18052–18056. https://doi.org/10.1073/pnas.0509478102

Norby,R.J.,Gu,L.,Haworth,I.C.,Jensen,A.M.,Turner,B.L.,Walker,A. P., …Winter, K. (2016). Informing models through empiricalrelationshipsbetweenfoliarphosphorus,nitrogenandphotosyn-thesisacrossdiversewoodyspeciesintropicalforestsofPanama.New Phytologist, 215(4), 1425–1437. https://doi.org/10.1111/nph.14319

Norby,R.J.,Warren,J.M.,Iversen,C.M.,Medlyn,B.E.,&McMurtrie,R. E. (2010). CO2 enhancement of forest productivity constrainedby limitednitrogenavailability.Proceedings of the National Academy of Sciences, 107(45), 19368–19373. https://doi.org/10.1073/pnas.1006463107

Oren,R.,Ellsworth,D.S.,Johnsen,K.H.,Phillips,N.,Ewers,B.E.,Maier,C.,…Katul,G.G.(2001).SoilfertilitylimitscarbonsequestrationbyforestecosystemsinaCO2‐enrichedatmosphere.Nature,411(6836),469–472.https://doi.org/10.1038/35078064

Ostertag, R. (2010). Foliar nitrogen and phosphorus accumulationresponses after fertilization: An example from nutrient‐limitedHawaiian forests. Plant and Soil, 334(1–2), 85–98. https://doi.org/10.1007/s11104‐010‐0281‐x

Pan,Y.,Birdsey,R.A.,Fang,J.,Houghton,R.,Kauppi,P.E.,Kurz,W.A.,…Hayes,D.(2011).Alargeandpersistentcarbonsinkintheworld’sforests.Science,333(6045),988–993.https://doi.org/10.1126/science.1201609

Poorter,H.,Bühler,J.,vanDusschoten,D.,Climent,J.,&Postma,J.A.(2012). Pot sizematters:Ameta‐analysis of the effects of rootingvolumeonplantgrowth.Functional Plant Biology,39(11),839–850.https://doi.org/10.1071/FP12049

Prada,C.M.,Morris,A.,Andersen,K.M.,Turner,B.L.,Caballero,P.,&Dalling,J.W.(2017).Soilsandrainfalldrivelandscape‐scalechangesin thediversityand functionalcompositionof treecommunities in

premontanetropicalforest.Journal of Vegetation Science,28(4),859–870.https://doi.org/10.1111/jvs.12540

Quesada,C.A.,Phillips,O.L.,Schwarz,M.,Czimczik,C.I.,Baker,T.R.,Patiño,S.,…Lloyd,J.(2012).Basin‐widevariationsinAmazonforeststructureand function aremediated by both soils and climate.Biogeosciences,9(6),2203–2246.https://doi.org/10.5194/bg‐9‐2203‐2012

Raaimakers, D., Boot, R. G. A., Dijkstra, P., Pot, S., & Pons, T. (1995).Photosynthetic rates in relationto leafphosphoruscontent inpio-neerversusclimaxtropicalrainforesttrees.Oecologia,102(1),120–125.https://doi.org/10.1007/BF00333319

Reich,P.B.,Hobbie,S.E.,Lee,T.,Ellsworth,D.S.,West, J.B.,Tilman,D.,…Trost,J.(2006).Nitrogenlimitationconstrainssustainabilityofecosystem response toCO2. Nature,440(7086), 922–925. https://doi.org/10.1038/nature04486

Santiago,L.S.,Korine,C.,Yavitt,J.B.,Harms,K.E.,Wright,S.J.,Garcia,M.N.,&Turner,B.L.(2011).Tropicaltreeseedlinggrowthresponsestonitrogen,phosphorusandpotassiumaddition.Journal of Ecology,100(2),309–316.https://doi.org/10.1111/j.1365‐2745.2011.01904.x

Schimel, D., Stephens, B. B., & Fisher, J. B. (2015). Effect of increas-ingCO2on the terrestrialcarboncycle.Proceedings of the National Academy of Sciences, 112(2), 436–441. https://doi.org/10.1073/pnas.1407302112

Slot,M.,Palow,D.T.,&Kitajima,K.(2013).SeedreservedependencyofLeucaena leucocephalaseedlinggrowthfornitrogenandphosphorus.Functional Plant Biology, 40(3), 244–250. https://doi.org/10.1071/FP12255

Steidinger, B. S., Turner, B. L., Corrales, A., & Dalling, J. W. (2015).Variability inpotential toexploit different soil organicphosphoruscompoundsamongtropicalmontanetreespecies.Functional Ecology,29(1),121–130.https://doi.org/10.1111/1365‐2435.12325

Tang,J.,Chen,J.,&Chen,X. (2006).Responseof12weedyspeciestoelevatedCO2inlow‐phosphorus‐availabilitysoil.Ecological Research,21(5),664–670.https://doi.org/10.1007/s11284‐006‐0161‐2

Thompson,J.B.,Slot,M.,Dalling,J.,Winter,K.,Turner,B.L.,&Zalamea,P.C.(2019).Datafrom:Species‐specificeffectsofphosphorusaddi-tionontropicaltreeseedlingresponsetoelevatedCO2. Dryad Digital Repository,https://doi.org/10.5061/dryad.777461k

Turner,B.L.,Brenes‐Arguedas,T.,&Condit,R.(2018).Pervasivephos-phorus limitation of tree species but not communities in tropicalforests.Nature,555(7696),367–370.https://doi.org/10.1038/nature25789

Vitousek, P. M. (1984). Litterfall, nutrient cycling, and nutrient lim-itation in tropical forests. Ecology, 65(1), 285–298. https://doi.org/10.2307/1939481

Vitousek,P.M.(2004).Nutrient cycling and limitation: Hawai'i as a model system.Princeton,NJ:PrincetonUniversityPress.

Vitousek,P.M.,&Howarth,R.W.(1991).Nitrogenlimitationonlandandinthesea:Howcanitoccur?Biogeochemistry,13(2),87–115.https://doi.org/10.1007/BF00002772

Winter, K., Aranda, J., Garcia, M., Virgio, A., & Paton, S. R. (2001).Effect of elevated CO2 and soil fertilization on whole‐plantgrowth and water use in seedlings of a tropical pioneer tree,Ficus insipida. Flora, 196(6), 458–464. https://doi.org/10.1016/S0367‐2530(17)30087‐7

Winter, K., Garcia, M., Gottsberger, R., & Popp, M. (2001). Markedgrowthresponseofcommunitiesoftwotropicaltreespeciestoel-evatedCO2when soil nutrient limitation is removed.Flora,196(1),47–58.https://doi.org/10.1016/S0367‐2530(17)30011‐7

Winter, K., & Lovelock, C. E. (1999). Growth responses of seedlingsof early and late successional tropical forest trees to elevated at-mospheric CO2. Flora, 194(2), 221–227. https://doi.org/10.1016/S0367‐2530(17)30900‐3

Wright, S. J. (2019). Plant responses to nutrient addition experimentsconducted in tropical forests. Ecological Monographs, https://doi.org/10.1002/ecm.1382

     |  11Functional EcologyTHOMPSON eT al.

Wright,S.J.,Yavitt,J.B.,Wurzburger,N.,Turner,B.L.,Tanner,E.V.J.,Sayer, E. J., … Corre,M. D. (2011). Potassium, phosphorus, or ni-trogen limit root allocation, tree growth, or litter production ina lowland tropical forest. Ecology, 92(8), 1616–1625. https://doi.org/10.1890/10‐1558.1

Würth,M.K.R.,Winter,K.,&Körner,C.H.(1998).In situresponsestoelevatedCO2intropicalunderstoreyplants.Functional Ecology,12(6),886–895.https://doi.org/10.1046/j.1365‐2435.1998.00278.x

Yang, X., Thornton, P. E., Ricciuto, D. M., & Hoffman, F. M. (2016).Phosphorus feedbacks constraining tropical ecosystem responsesto changes in atmospheric CO2 and climate.Geophysical Research Letters,43(13),7205–7214.https://doi.org/10.1002/2016GL069241

Zalamea,P.,Turner,B.L.,Winter,K.,Jones,F.A.,Sarmiento,C.,&Dalling,J.W.(2016).Seedlinggrowthresponsestophosphorusreflectadultdistributionpatternsoftropicaltrees.New Phytologist,212(2),400–408.https://doi.org/10.1111/nph.14045

Ziska, L.H.,Hogan, K. P., Smith, A. P.,&Drake, B.G. (1991).Growthand photosynthetic response of 9 tropical species with long‐term

exposure to elevated carbon‐dioxide. Oecologia, 86(3), 383–389.https://doi.org/10.1007/BF00317605

SUPPORTING INFORMATION

Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle.

How to cite this article:ThompsonJB,SlotM,DallingJW,WinterK,TurnerBL,ZalameaP‐C.Species‐specificeffectsofphosphorusadditionontropicaltreeseedlingresponsetoelevatedCO2. Funct Ecol. 2019;00:1–11. https://doi.org/10.1111/1365‐2435.13421