Effects of Amazonian Dark Earths on growth and leaf nutrient balance of tropical tree seedlings

15 Pages·2015·0.88 MB·English

PlantSoil(2015)396:241–255
DOI10.1007/s11104-015-2558-6
REGULARARTICLE
Effects ofAmazonian Dark Earths on growth and leaf nutrient
balance of tropical tree seedlings
EstelaQuintero-Vallejo&MarielosPeña-Claros&
FransBongers&MarisolToledo&LourensPoorter
Received:26December2014/Accepted:7June2015/Publishedonline:10July2015
#TheAuthor(s)2015.ThisarticleispublishedwithopenaccessatSpringerlink.com
Abstract ADE,buttheyinvestedinleavesandleafareainADE,
Background and aims Amazonian Dark Earths (ADE) althoughthisdidnotleadtofastergrowthrate.Species
are ancient anthropogenic soils distributed in the respondeddifferentlytosoilCaincrement;somespecies
Amazonbasin.Theyarecharacterizedbyhighnutrients seemed to suffer from Ca toxicity as indicated by low
such as phosphorus, calcium, potassium and nitrogen. survival,othersfromnutrientimbalance,whereasother
We studied the effect of ADE on growth, morphology species increased their leaf calcium, phosphorus and
and physiology of 17 tree species from a Bolivian nitrogen concentration in ADE. Only for this latter
tropicalmoistforest. group of nutrient accumulators, there was a positive
Methods Weconductedagreenhouseexperimentwhere interspecific relationship between leaf Ca and seedling
seedlingsweregrownfor2–4monthsonADEandnon- growthrates.
ADE. We evaluated soil nutrient concentrations, seed- Conclusions ADE did not lead to increased seedling
ling growth, leaf and root functional traits, and leaf growth. The ability of plants to colonize patches of
nutrientconcentrations. ADEmightdependonplantresponsestoincreasedsoil
Results Soiltypeaffected10outof24evaluatedseed- Caandtheircapacitytoregulateinternaltissuecalcium
lingtraits.Seedlingsdidnotinvestmoreinrootsinnon- tobalancenutrition.
ResponsibleEditor:HansLambers. Keywords AmazonianDarkEarths.Bolivia.Calcium
Electronicsupplementarymaterial Theonlineversionofthis copingstrategies.Nutrientimbalance.RelativeGrowth
article(doi:10.1007/s11104-015-2558-6)containssupplementary Rate
material,whichisavailabletoauthorizedusers.
: : :
E.Quintero-Vallejo(*) M.Peña-Claros F.Bongers
L.Poorter
Introduction
ForestEcologyandForestManagementGroup,Wageningen
UniversityandResearchCentre,P.O.Box47,6700
AAWageningen,TheNetherlands Nutrient availability in tropical soils is the result of
e-mail:[email protected] complex biogeochemical processes that generate high
: : spatialheterogeneity (Townsend etal. 2008).This het-
E.Quintero-Vallejo M.Peña-Claros M.Toledo
erogeneityisimportant,sinceitcandeterminetreespe-
InstitutoBolivianodeInvestigaciónForestal(IBIF),P.O.Box
6204,SantaCruzdelaSierra,Bolivia cies distribution at different spatial scales (John et al.
2007). However, the exact mechanisms that drive the
M.Toledo
underlyingpatternsofplantdistributionarestillfarfrom
FacultaddeCienciasAgrícolas,UniversidadAutónoma
beingfullyunderstood.Thislackofexplanationispartly
GabrielRenéMoreno,Km9CarreteraalNorte,SantaCruzde
laSierra,Bolivia becauseplantsresponddifferentlytoamultitudeofsoil-
242 PlantSoil(2015)396:241–255
related factors, such as water availability (Brenes‐ elements such as Ca and magnesium (Mg), causing
Arguedas et al. 2013) and nutrients (Ordoñez et al. reduced development of roots, depression of growth
2009). Among nutrients, nitrogen (N) and phosphorus and potentially plant death (George et al. 2012). On
(P)arethemajorlimitingelementsforprimaryproduc- the other hand, in soils with high pH (~7.4) such as
tivity of forests (Vitousek et al. 2010). Nevertheless, calcareous soils, increased concentrations of available
recent fertilization studies have found that potassium Ca can react withinorganic P and form calcium phos-
(K)andcalcium(Ca)influenceplantgrowthintropical phate,whichmakesPunavailableformostplants(Tyler
forest,insimilarwaysasNandP(Baribaultetal.2012; 1996). Besides, in soils with high pH, some elements
Wrightetal.2011). such as Fe, Mn, Zn and Cu form oxide-bound com-
Plantsrespondtosoilnutrientavailabilitythrougha plexes and become unavailable, thus limiting plant
suite of traits, by adjusting their allocation patterns, growth(Lee1998).
morphology, tissuechemistry and physiology. With an In general, soils in the tropics have low pH (Von
increaseinsoilnutrientavailability,plantsshiftbiomass Uexküll and Mutert 1995) and a low availability of a
allocationtoaerialstructures,suchasleaves,insteadof rangeofnutrients.However,Neotropicalsoilscanalso
underground structures such as roots (Poorter et al. be diverse in their chemical properties as a result of
2012).Thisisusuallyaccompaniedbychangesinother different geological processes. In particular, the
plant traits such as an increase in specific leaf area Amazon basin has experienced a long and dynamic
(SLA), N and P concentration in leaves, and growth geological history (Hoorn et al 2010). In combination
(Ordoñezetal.2010;Ordoñezetal.2009).Ontheother withdifferentbiogeochemicalprocesses(Quesadaetal.
hand,insoilswithlow-nutrientavailability,plantsgrow 2010)thishasledtoawidevariationinsoilcharacter-
slowly, haveasmaller leafareaandallocatemorebio- istics and different availability of nutrients such P, Ca,
masstoroots,thusenhancingnutrientuptake(Baraloto Mg, and K. Despite the fact that most of the Amazon
et al. 2006). Plant tissue chemistry changes with soil basin has a low soil fertility, there are patches of
fertility, which is usually reflected in the nutrient con- Amazonian Dark Earths (ADE) with high soil fertility
centrationofleaves,particularlyleafNandPandtheir that are the product of past human inhabitation of the
ratio (Ågren 2004; Townsend et al. 2007). Apart from region since pre-Columbian times (Glaser and Birk
soilfertility,foliarnutrientconcentrationsareshapedby 2012).
a number of other factors, such as soil pH (Lambers AmazonianDarkEarthsarecharacterizedbyathick
et al. 2008; Viani et al. 2014), plant demands for ele- darkorgraytoplayer(upto1.5mdeep)withpresence
mentssuchascalcium(Ca)toincreasetissuetoughness of ceramics that indicates past indigenous settlements
(Baribault et al. 2012), negative interactions between (Sombroek1966;WoodsandGlaser2004).ADEhavea
tissue P and aluminum (Al) when plants tend to accu- blackcolorduetohighconcentrationsoforganicmatter
mulate Al (Metali et al. 2015), species phylogeny andblackcarbon,theyhavehighPandCa,andhigher
(Watanabe et al. 2007), and symbiotic relationships pH than the natural soils commonly found in the
between plants and microorganisms (Nasto et al. Amazon, such as Ultisols and Oxisols (Falcão et al.
2014). As a result, plant nutrient concentrations and 2009; Glaser and Birk 2012; Sombroek 1966). ADE
theirstoichiometrycanbeuncoupledfromthoseinthe seem to be the product of additions of charcoal and
soil. organicwastewhichchangesoilphysic-chemicalchar-
Soil pH is an important factor in plant nutrition; it acteristics(Schmidtetal.2014).Forexample,charcoal
maylimittheavailabilityofarangeofelementsthatare addition leads to a more stable organic matter content,
important for plant growth (e.g., iron [Fe] and zinc andahighsoilnutrientretentioncapacity(Glaseretal.
[Zn]),and may raise the availability ofotherelements, 2002),whiletheadditionoforganicwasteleadstohigh
such as Al, that could cause plant mortality (Lambers concentrationsofP,N,andCa(Glaser2007).Compared
etal.2008).Forexample,insoilswithalowpH(<5.8), with natural Central Amazonian soils, ADE are more
increasesinAlandFeavailabilityleadtodecreasesinP fertile because ofthe combination of high pHand Ca-
availabilitybyreactionofAlandFewithinorganicPto and P- concentrations. For this reason, ADE are often
form plant unavailable insoluble compounds used by local farmers to grow crops faster for longer
(McDowelletal.2003).Furthermore,anincreaseinsoil periods oftime compared withslash and burn agricul-
Al affects root elongation or plant uptake of other turepracticedonregularsoilsintheAmazon(German
PlantSoil(2015)396:241–255 243
2003).Similarly,aforestgrowingonADEhasincreased area and leaf area ratio, which are traits that enhance
gross primary productivity, net primary productivity, light capture and carbon gain. We also predict that
andhigherratesoffinerootproductioncomparedwith seedlings growing on less fertile non-ADE will have
aforestgrowingonnon-ADE(Doughtyetal.2014). higherbiomassallocationto(secondary)rootsandpro-
Bystimulatingproductivity,thehighfertilityofADE duce roots with higher specific root length to enhance
could promote changes in plant traits associated with nutrientuptake.
acquisitionstrategiessuchashigherrelativegrowthrate,
higher leaf biomass and higher nutrient concentrations
intheleaves.Evaluatingchangesinplanttraitsassoci- Methods
atedwithADEprovidesauniqueopportunitytounder-
standtreeresponsestosoilfertilityasaconsequenceof Speciesandcollectionsite Seventeentreespecieswere
pasthumaninhabitationandsoilmodification.Itcould selected for the study based on seed availability.
also help to understand how soil fertility influences Hereafter species will be referred by the genus name
species composition in habitats that do not vary in (Table1).Mostoftheseedswerecollectedfromasemi-
climaticconditionsandwateravailability. evergreen forest (La Chonta forest concession) located
Herewepresentresultsofagreenhouseexperiment, intheprovinceofGuarayos,Bolivia(15°47′S,62°55′
in which seedlings of 17 tropical tree species were W) in July and August of 2012. The site has a mean
grown onADEand non-ADE. We examinedhow soil annualtemperatureof24.3°Candmeanannualrainfall
fertilityofADEaffectedseedlinggrowth,biomassallo- of1580mm(datafrom2000to2006fromLaChonta)
cation,andmorphologicalandchemicaltraitsoftropical withadryseason(<100mm/month)fromMaythrough
tree species. We hypothesized that plants growing on September(Peña-Clarosetal.2012).SoilsatLaChonta
ADE would have higher growth rates compared with are a mosaic of poor soils from the Brazilian
plantsgrowingonnon-ADE.Wealsohypothesizedthat Precambrian Shield (Navarro and Maldonado 2002)
thehighergrowthratesofseedlingsonADEwouldbe andmorefertilesoilsthataretheproductofsedimenta-
associatedwithincreasesinbiomassallocationtostem tion and erosion originated from the uplift of central
andleaves,leafN,CaandPconcentrations,specificleaf Andes (Latrubesse et al. 2010). A previous study
Table1 Scientificnamesand
familiesoftreespeciesusedfor ScientificName Family Nodulespresentinroots
thisstudy
Reported Observed
Albizianiopiodes(SpruceexBenth.)Burkart Fabaceae +1 +
Anadenantheracolubrina(Vell.)Brenan Fabaceae +3 −
Enterolobiumcontortisiliquum(Vell.)Morong Fabaceae +2,3,4 −
Erythrinacrista-galliL. Fabaceae +4 +
HymenaeacourbarilL. Fabaceae −2,5 −
MachaeriumvillosumVogel Fabaceae +2 −
Informationonpresenceofnod-
ulesinroots,asreportedinthe OrmosianobilisTul Fabaceae +1 −
literature,isshownwiththe PoeppigiaproceraC.Presl Fabaceae −
numbercorrespondingtoitsref-
Samaneatubulosa(Benth.)Barneby&J.W.Grimes Fabaceae +
erence.Informationonobserved
Schizolobiumparahyba(Vell.)S.F.Blake Fabaceae
nodulesinrootsofseedlingsused
inthisexperimentisshown.(+) Terminaliaoblonga(Ruiz&Pav.)Steud. Combretaceae
representsnodulespresent,and CarinianaianeirensisR.Knuth Lecythidaceae
(−)representsnodulesabsent
ChorisiaspeciosaA.St.-Hil. Malvaceae
1(deFariaanddeLima1998)
GuazumaulmifoliaLam. Malvaceae
2(Barberietal.1998)
CedrelafissilisVell. Meliaceae
3(SprentandParsons2000)
SwieteniamacrophyllaKing Meliaceae
4(Frionietal.2001)
SapindussaponariaL. Sapindaceae
5(Limaetal.2006)
244 PlantSoil(2015)396:241–255
showedthatLaChontaalsohasAmazonianDarkEarth dilutetheeffectofsoilnutrients,weconsideredthatthis
(ADE),whichareanthropogenicsoilsindicatinginhab- dilution might not affect our experiment because the
itation of this forest in the past (Paz-Rivera and Putz differencesbetween ADEand non-ADE wereretained
2009). for most of the soil variables except total N (Table 2).
Chemical and physical characteristics of the original
Study site The experiment was carried out in a shade soilsandofthemixturewithsandwereassessedinthe
house at the Agronomy Faculty of the Universidad Centro de Investigación Agrícola Tropical (CIAT), in
Autónoma Gabriel Rene Moreno in Santa Cruz de la Santa Cruz, Bolivia (protocols for chemical and
Sierra, Bolivia (16° 30′ S, 63° 10′ W). The climate in physicalpropertiescanbeseeninOnlineResource 2).
thiscityiswarm,withanaveragetemperatureof24.2± ThemaindifferencesinsoilvariablesbetweenADEand
0.8 ° C, and mean annual rainfall of 1308±391 mm non-ADEsandmixeswerefoundforCaconcentration,
(Data from National Institute of Meteorology, Bolivia; P concentration, total exchangeable base and cation
1949–2013). The 6×3 m shade-house was built and exchangecapacity(ratiosbetweenADEandnon-ADE
coveredwithblackmesh(2×1.5mm)togiveanirradi- were>2inallcases);soilpHwasslightlyalkaline(>7)
ance of about 10 % of full sunlight (around 400 lux). in both soils (ADE: 7.8; non-ADE:7.6) (Online
Thisleveloflightissimilarorevenhigherthanthelight Resource1).
encountered in the understory of a tropical rain forest FromApriltoSeptemberof2013seedsweregermi-
(~1–2%),whereseedlingsofmosttropicaltreespecies nated in trays of 25×30 cm filled with either ADE or
growwell(Poorter1999). non-ADE sand mixes. When seedlings produced their
first leaves, eight seedlings per species per soil type
Soils and shade-house experiment The soil used for were harvested and 12 seedlings per species per soil
testing germination and plant growth was collected at typeweretransplantedtopotsof650ml(9.2cmdiam-
La Chonta, where we previously identified a site with eter×15 cm height). The transplanted seedlings were
darkearthandthepresenceofpotteryshards,hereafter placed in a completely randomized design in the
ADE, and a site at least 500 m away with more clear shade-houseandwerere-randomizedevery3weeksto
soils, hereafter non-ADE. Soils were mixed with river ensure that all seedlings were growing under similar
sandataratioof4:1(w/w),toallowadequatedrainage light and temperature conditions. Plants were watered
and to facilitate harvesting of the whole root system, dailyoreveryotherdaydependingontheweather,since
including fine roots. Although the use of sand could the shade house allowed rain water to pass through.
Table2 Soilcharacteristicof
AmazonianDarkEarthsandnon- ADE non-ADE ADE/River non-ADE/ Ratiovariableson
AmazonianDarkEarthcollected sand Riversand ADE:non-ADE
inLaChontaforestconcession,
beforeandaftermixingthemwith pH 7.4 7 7.8 7.6 1
riversand Electricconductivity(μS/cm) 226 108 218 149 1.5
Ca(cmol/kg) 12.2 4.8 10 4.2 2.4
Mg(cmol/kg) 1.3 1 1.1 0.8 1.4
Na(cmol/kg) 0.11 0.04 0.09 0.05 1.8
K(cmol/kg) 0.38 0.26 0.28 0.2 1.4
Olsen-P(mg/kg) 36 12 35 11 3.2
TotalN(%) 0.31 0.16 0.18 0.11 1.6
TotalExchangablebases 14 6.1 11.5 5.3 2.2
CationExchangeCapacity 14 6.1 11.5 5.3 2.2
RatiosbetweenvaluesonADE
BaseSaturation(%) 100 100 100 100 1
andnon-ADEafterthemixing
OrganicMater(%) 3.3 2.7 2.9 2.2 1.3
withriversandarepresented.
Phosphorusreportedonthistable Sand(%) 49 57 68 73 0.9
wasobtainedusingOlsenMethod Silt(%) 39 36 25 20 1.2
forsoilPextraction(See,Online Clay(%) 12 7 7 7 1
Resource1)
PlantSoil(2015)396:241–255 245
Once seedlings were established, at around half of the Wemeasuredstemlengthfromthebasetotheapical
growingperiod,thenumberofleavesoneachseedling bud,andtopandbasestemdiameter.Stemvolumewas
wascountedandthenewestfullydevelopedleaveswere calculated using the formula for a section of a cone.
tagged with colored threads. At the time of harvesting Lamina and petioles, roots and stems were stored in
wecountedthenumberofnewlyproducedleaves. separatepaper bags, oven-driedat70°Cfor 72h,and
Asecondharvestwasdonebetween2and4months weighted again to determine dry weight of plant sec-
after transplanting, depending on the species. tions. Using the dry weights we calculated root mass
Differences in length of the growing period were due fraction(RMF, g g−1),leafmassfraction(LMF, g g−1)
to differences in germination time, and lack of seed and stem mass fraction (SMF, g g−1) as the weight of
availability of some species at the time we started the eachplantpartoverthetotalseedlingdrymass.Wealso
experiment. At each harvest, we determined the fresh calculated stem, root and leaf dry matter content
massofroots,stems,andleaves;leavesweresectioned (SDMC,RDMC,LDMC,respectively,ingg−1)asthe
into the leaf lamina and the petiole, or rachis for dry mass over the fresh mass of that section. Specific
compoundleaves,andsectionswereweighedseparate- leaf area (SLA, m2 Kg−1) was calculated as the ratio
ly. Laminae were scanned with a desktop scanner between leaf area and leaf dry mass, leaf area ratio
(Cannon LIDE 20, Canon, USA) and leaf area was (LAR,m2Kg−1)asthetotalleafareaovertotalseedling
determined using the software program ImageJ dry mass, specific root length (SRL, m Kg−1) as root
(Rasband2008).Leafresistancetomechanicaldamage lengthoverdryrootmassandrootlengthperplantmass
was measured for nine of thirteen species using force (RLPM m Kg−1) as the root length over total seedling
to punch. Force to punch was determined on one or drymass.
two of the newest expanded leaves using a field pen- These traits determine important functional charac-
etrometer,whichconsistedofaflat-endnailof3.2mm teristics for the plants. Biomass fractions of seedling
indiameter;thenailwasattachedtotheinnerpartofa sections describe how plants allocate biomass to light
syringe and a water basin on top. A leaf was placed intercepting tissue in the case of leaves, or nutrient
between two acrylic plates both having a 6 mm diam- capturingtissueinthecaseofroots.Stem,rootandleaf
eter hole. The holes were located in the same position dry matter content, and force to punch indicate tissue
so that the nail could cross the leaf between the two toughness, which are thought to be good proxies for
plates similar to Aranwela et al. (1999). Force to tissuelongevity.SLAandLARindicatehowefficiently
punchtheleafwasdeterminedbytheweight(convert- plantsinvestinlightinterception.MeasurementsofSRL
edtoNewton)ofthewateraddedtothebasinthatwas andRLPMindicatehowthebiomassthatisallocatedto
necessary to penetrate the leaf, divided by the circum- roots can be efficient in nutrient capture through in-
ference of the punch nail. crease of absorption surface (Markesteijn and Poorter
Root length (RL) was determined following 2009).
Newman(1966).Rootswereplacedinatrayfilledwith Leaf nutrient N,P, and Ca concentration was deter-
water. The tray was covered by a transparent sheet minedforthreerandomlyselectedseedlingsperspecies
marked with 1×1 cm square grid and the number of persoiltype.Forfivespecies(Albizia,Anadenanthera,
intersections between roots and the gridlines were Cedrela, Machaerium and Samanea) we selected a
countedhorizontallyandvertically.Thereafter,totalroot seedling that had a leaf dry weight>50 mg, the mini-
lengthwas determined as R=πNA/2H, where R is the mum quantity required for leaf tissue analyses in the
total length ofthe root (cm), N isthe number ofinter- laboratory.Extraction ofN, PandCaweredoneusing
sectionsbetweentherootandthegridlines,Aisthearea digestion with H SO , Se, and salicylic acid
2 4
of the rectangle (cm2) and H is the total length of the (Novozamsky et al. 1983). After digestion, N and P
straight lines of the grid (cm) (Newman 1966). Root contentwasmeasuredwithacontinuous-flowanalyzer
morphology was described by the presence/absence of (SANPlusSegmentedFlowAnalyserSkalarSA-4000,
taprootsandbythenumberofsecondaryroots.When Skalar UK, York, UK), and Ca content was measured
tap roots were present both the diameter of the base using an Atomic Absorption Spectrometer (AAS-
(closetothetransitiontostem)andthediameterofthe Varian Spectra AA-600, Varian, Palo Alto, USA).
tip were measured. The diameter of three secondary Chemical tissue analyses were done at the facilities of
rootswasmeasuredwithadigitalcaliper. WageningenUniversity,TheNetherlands.
246 PlantSoil(2015)396:241–255
Data analyses We quantified survival percentage for P. Using the information of all species, we performed
eachspeciesbydividingthenumberofseedlingsalive linear regressions to test the relationship between the
at the time of the second harvest by the number of differenceinleafCa(Ca )andthedifference
ADE-nonADE
transplanted seedlings. We compared survival percent- in leaf P (P ) as well as the relationship be-
ADE-nonADE
agebetweensoiltypesusingachi-squaretest.Relative tweenthedifferenceinleafCa(Ca )andthe
ADE-nonADE
height growth rate (RGRh) was calculated for each differenceinleafN(N ).
ADE-nonADE
seedling as: RGRh¼lnh2−lnh1, where h represents the Based on species leaf P and N responses to soils
t2−t1 n type(see results y axes; Fig. 4a and b), we separated
seedling height at time n; t is time at transplanting, t
1 2
the species into the following groups: species whose
time at harvesting, and (t - t ) is the number of days
2 1
differences in leaf P (P ) were positive, im-
betweentransplantingandharvest.RGRbasedonplant ADE-nonADE
plyinghigherleafPonADE(hereafterreferedaspos-
biomass (RGRb) was calculated for each species using
itive for P ); species whose differences in P (P
the information of both harvests as: RGRb¼lnM2−lnM1, ADE-
t2−t1 nonADE) were negative, implying lower leaf P on ADE
wherelnMn istheaverageofthenaturallogofthedry (negative for P); species whose differences in leaf N
massofeachseedlingatharvestn. (N )werepositive,implyinghigherleafNon
ADE-nonADE
Continuousvariableswerelog10-transformed,whereas ADE(positiveforN);specieswhosedifferencesinleaf
ratios were arcsine-transformed prior to statistical analy- N(N )werenegative,implyinglowerleafN
ADE-nonADE
ses.TotestfortheeffectofADEandnon-ADEontotal on ADE (negative for N). Using these four response
biomass,RGRh,andplanttraits,weperformedatwo-way groups, we tested how Ca uptake was related with
ANOVA, with soil type and species as factors. For this RGRb.WeevaluatedCabecauseitwasthenutrientin
analysisweused13speciesinsteadof17becausewehad the leaves that consistently increased in ADE for all
significantmortalityinfourspeciesthatleftuswithfew species (Fig. 3a). Therefore, we performed a linear
replicatespertreatmenttomakecomparisonsbetweenthe regression for each response group between leaf Ca
two soils. We used an unbalanced design because two concentrationinleavesastheindependentvariableand
species(CarinianaandHymenaea)haddifferentnumber theaverageRGRbasthedependentvariable.Finally,we
of seedlings per soil type. The interaction between soil alsotestedhowCauptakewasrelatedwithleafPuptake
type and species was also evaluated. For all traits but or leaf N uptake for each response group using linear
RGRh,dataofthesecondharvestwereused.Theamount regressions. All statistical analyses were done using
ofvariancethatwasexplainedbyeachfactorwasdeter- Genstat16thed.(VSNInternationalLtd).
mined by dividing the sum of squares associated with
each factor (soil or species) and their interaction by the
totalsumofsquaresofthemodel.Aseparateanalysiswas
doneforeachspeciesusingone-wayANOVA,withsoil Results
typeasfactor.Theeffectofsoiltype onleafproduction
ratewastestedusinganon-parametricKruskal-Wallistest Eight out of 17 species showed some mortality during
becausethedatawerenotnormallydistributed. theexperiment.TerminaliaandPoeppigiahadasignif-
We detected that the percentage of leaf Ca was al- icantly higher mortality in ADE than in non-ADE soil
wayshigheronADE,whereasthepercentagesofleafP (χ2:10.83, p<0.05; χ2:12.08, p<0.05, respectively),
andleafNweresimilarorloweronADEthaninnon- whereas Schizolobium had a significantly higher mor-
ADE,despitethehigherconcentrationoftheseelements talityinnon-ADE(χ2:15.41,p<0.05)(Fig.1).
in ADE (Table 1; see results; Fig. 3). Therefore, we
exploretherelationshipsbetweenpairsoftissuenutrient PlantresponsestoADE
concentration(CaandP;CaandN)fromeachsoil.We
averagedtheCa,NandPconcentrationintheleavesof Wefoundsignificantdifferencesbetweensoiltypesfor
seedlingsgrowinginADEandnon-ADE,andcalculat- ten out of 24 traits, but the proportion of the variance
ed the difference between these amounts in each soil explainedbysoilsforeachofthesevariableswasvery
t(cid:1)ype per species, as follows: D(cid:3)ifferenceADE−nonADE ¼ low(lessthan2.8%,Table3).Specieshadasignificant
Nutrienti;ADE−Nutrienti;non−ADE ,whereNutrientstands effect on all variables, explaining 35 to 96 % of the
fortheconcentrationintheleaf,andistandsforCa,Nor variance(Table3).
PlantSoil(2015)396:241–255 247
Fig.1 Seedlingsurvivalof17 ADE
speciesoftropicaltreesin
non-ADE
AmazonianDarkEarths(ADE) 100
andnon-AmazonianDarkEarth
(non-ADE)soilsafterthe %)
(80
growingperiodinthegreenhouse al
v
vi
r60
u
S
40
20
0
WeexpectedsoilfertilityfromADEtohaveaposi- Enterolobiumseedlingshadheavierstemsinnon-ADE
tiveeffectonplantgrowth,butmostofthespecieshada than in ADE (Online Resource 2). Specific leaf area
similar growth rate on both soils. Total biomass (SLA,ameasureofareainvestmentperleafmass)and
(Fig. 2a), height growth (RGRh, Fig. 2b) and biomass leafareainvestmentperplantmass(LAR)weresignif-
growth(RGRb)didnotdiffersignificantlybetweensoil icantlyhigherforseedlingsinADE(Table3).Wefound
types (Table 3). However, the interaction between soil a significant interaction term for specific root length
type and species was significant for total biomass and (SRL) (a measure of plant investment in root length)
RGRh (Table 3), indicating that some species grew (Table 3); Enterolobium seedlings had a significantly
faster in ADE, whereas other species grew faster in higher SRL in ADE, whereas Swietenia seedlings had
non-ADE. SeedlingsofCedrela and Chorisia had sig- higherSRLinnon-ADE(OnlineResource2).Leafand
nificantly more biomass in ADE (one-way ANOVA: rootdrymasscontentweresignificantlyhigherinnon-
CedrelaF=7.29,p=0.013;ChorisiaF=5.40,p=0.03), ADE(Table3).LeafCaconcentrationwassignificantly
whereasMachaeriumhadasignificantlyhigherbiomass higher in ADE (Table 3) this was particularly the case
in non-ADE (one-way ANOVA, F=4.87, p=0.039) for Ormosia, Erythrina, Albizia, andCariniana
(Fig. 2a). Furthermore, Swietenia had a higher RGRh (Fig. 3a). Leaf N concentration did not differ signifi-
in ADE (one-way ANOVA, F=9.04, p=0.006) cantlybetweensoiltypes(Table3,Fig.3b).Wefounda
(Fig. 2b), whereas Chorisia had a significantly higher significantinteractionbetweensoilandspeciesforleaf
RGRh in non-ADE (one-way ANOVA, F=18.88, P concentration (Table 3), caused by a significantly
p<0.001). higher leaf P concentration in ADE for Machaerium
Overall,leafproductionratewassimilarforbothsoil (one-wayANOVA,F=8.36,p=0.04)andinnon-ADE
types (Kruskal-Walis H=2.97; p=0.084), although for Erythrina (one-way ANOVA, F=128.2, p<0.001)
some species such as Cedrela, Chorisia, and (Fig.3c).
Cariniana had a higher leaf production rate in ADE
than in non-ADE, whereas Swietenia had higher leaf Nutrientstoichiometryinrelationwithgrowthrates
production rate in non-ADE soil (Online Resource 2).
Wefoundthatbiomassallocationtoleaves(LMF)was We explored the differences in leaf nutrient concen-
significantlyhigherinADEthaninnon-ADE(Table3). tration between seedlings growing in ADE and non-
Therewerenosignificantdifferencesinbiomassalloca- ADE soils. We found that all species had a higher
tion to roots (RMF) between soil types. There was a leaf Ca in ADE than non-ADE (positive values on x
significantinteractionbetweensoilsandspeciesinallo- axis, Fig. 4a), whereas species had either a higher
cation to stems (SMF) (Table 3). Seedlings of (positive values on y axis, Fig. 4a) or a lower leaf P
Anadenanthera had heavier stems in ADE (F=14.71, inADEthaninnon-ADE(negativevalueson yaxis,
p<0.001) than in non-ADE, whereas Erythrina and Fig. 4a). This resulted in a significant negative
248 PlantSoil(2015)396:241–255
Table3 Averagesofmorphological,physiologicalandchemicalvariablesmeasuredonseedlingsof13speciesoftropicaltreesgrowingon
ADE(n=12seedlingsperspecies)andnon-ADEsoils(n=12seedlingsperspecies)
Trait Soil Species SoilxSpecies
ADE N-ADE R2 F p SSs/SSt F p SSspp/SSt F p SSsxspp/SSt
Biomass(g) 0.932 0.94 0.89 0.39 0.531 0.0002 184.11 <.001 0.88 3 <.001 0.014
RGRh(mm.mm.d−1) 0.043 0.042 0.87 0.09 0.767 0.0000 147.47 <.001 0.84 6.53 <.001 0.037
RGRb(g.g.d−1) 0.016 0.016 – 0.005 0.944 0.0000 – – – – – –
RootMassFraction(RMF)(gg−1) 0.232 0.244 0.68 3.36 0.068 0.0040 46.54 <.001 0.67 0.58 0.859 0.008
StemMassFraction(SMF)(gg−1) 0.255 0.261 0.67 1.8 0.181 0.0022 43.4 <.001 0.64 2.15 0.015 0.032
LeafMassFraction(LMF)(gg−1) 0.513 0.494 0.61 8.06 0.005 0.0117 33.19 <.001 0.58 1.14 0.331 0.020
Diameterofsecondaryroots(mm) 0.481 0.49 0.61 0.08 0.776 0.0001 31.74 <.001 0.59 1.24 0.253 0.023
Secondarytoprimaryrootmass(gg−1) 0.897 0.755 0.38 2.98 0.086 0.0069 12.57 <.001 0.35 1.05 0.401 0.029
Numberofsecondarysoots 29.17 30.66 0.53 1.43 0.233 0.0026 22.77 <.001 0.49 1.66 0.076 0.036
SpecificRootLength(mKg−1) 17,860 15,310 0.57 0.36 0.549 0.0006 26.88 <.001 0.52 2.69 0.002 0.052
RootLengthRatio(mKg−1) 98,740 74,760 0.78 0.11 0.736 0.0001 79.19 <.001 0.77 1.74 0.059 0.017
RootDryMatterContent(RDMC)(gg−1) 0.173 0.187 0.78 15.91 <.001 0.0128 78.82 <.001 0.76 0.78 0.672 0.008
StemDryMatterContent(SDMC)(gg−1) 0.282 0.293 0.91 7.81 0.006 0.0027 218.1 <.001 0.90 2.12 0.016 0.009
LeafDryMatterContent(LDMC)(gg−1) 0.27 0.279 0.96 14.98 <.001 0.0022 551.38 <.001 0.96 0.98 0.464 0.002
StemDensity(gcm−3) 0.31 0.316 0.90 1.6 0.207 0.0006 188.95 <.001 0.89 1.93 0.031 0.009
SpecificLeafArea(SLA)(m2Kg−1) 41.66 39.46 0.74 4.8 0.029 0.0046 63.23 <.001 0.73 0.83 0.624 0.010
LeafAreaRatio(LAR)(m2Kg−1) 18.83 17.27 0.61 9.54 0.002 0.0140 32.52 <.001 0.57 1.17 0.306 0.021
ForcetoPunch(Nmm−1) 0.229 0.244 0.69 4.31 0.039 0.0065 49.62 <.001 0.68 0.88 0.543 0.012
LeafThickness(mm) 0.123 0.124 0.90 1.01 0.315 0.0005 202.28 <.001 0.89 1.87 0.058 0.008
LeafCaconcentration(mgg−1) 19.2 16.8 0.95 26.52 <.001 0.0278 71.3 <.001 0.90 1.67 0.102 0.021
LeafNconcentration(mgg−1) 34.7 35.4 0.95 1.35 0.251 0.0012 88.88 <.001 0.94 0.87 0.579 0.009
LeafPconcentration(mgg−1) 1.9 2.1 0.90 6.78 0.012 0.0134 32.42 <.001 0.77 5.02 <.001 0.119
Ca:P 11.06 9.95 0.83 3.74 0.059 0.0124 18.73 <.001 0.74 1.79 0.075 0.071
Ca:N 0.59 0.51 0.94 22.58 <.001 0.0258 65.31 <.001 0.90 1.41 0.191 0.019
N:P 20.51 19.84 0.83 0.41 0.522 0.0013 19.86 <.001 0.77 1.52 0.146 0.059
Theoverallr2 ofthemodelandtheANOVAresultsareprovided.Theeffectofsoil,speciesortheinteractionispresentedwiththeirFvalues
andtheirsignificancelevel.Thepercentageofvariationexplainedbyeachfactor(SS /SS )ortheinteractionwascalculatedastheratio
factor total
betweenthefactorsumofsquaresorinteractionsumofsquaresandtotalsumofsquares;thus,variationexplainedbysoils(SS/SS),
s t
variationexplainedbyspecies(SS /SS),variationexplainedbytheinteractionbetweensoilandspecies(SS /SS)
spp t sxspp t
relationship between the difference in leaf Ca con- difference in leaf P concentration (P )
ADE-nonADE
centration (Ca ) and the concomitant (Fig. 4a). A similar trend was found for leaf Ca
ADE-nonADE
a
b
3.50 0.25 ADE non-ADE
(g)23..5000 -1)m d 0.20 *
s m 0.15
s2.00
oma1.50 mm. 0.10 *
Bi1.00 * * h(
0.50 * GR 0.05
0.00 R 0
Fig.2 Biomassatfinalharvest(a)andrelativeheightgrowthrate (open bars). Means and standard errors are shown. Asterisks
(RGRh) (b) of seedlings of 13 species growing in Amazonian indicatesignificantdifferencesbetweensoilsatthespecieslevel
DarkEarth(ADE)(blackbars)andnon-AmazonianDarkEarths (p≤0.05)
PlantSoil(2015)396:241–255 249
a b
45 5
40 4.5 * ADE non-ADE
4
35
gg)-1 30 * -1g g) 3.35
eaf Ca(m 122505 * * * LeafP (m 12..255 *
L 10 1
0.5
5
0
0
c
55
50
45
-1g) 40
35
g
m 30
(
N 25
af 20
Le 15
10
5
0
Fig.3 Leafcalcium(a),phosphorus(b)andnitrogen(c)concen- soil type.Speciesthataremarkedwithablackpoint(•)areN-
trationsofseedlingsof13tropicaltreespeciesgrowinginAma- fixinglegumesreportedonTable1.Barsrepresenttheaverageper
zonianDarkEarthssoils(blackbars)andnon-AmazonianDarks soiltype,withstandarderrors.Asterisksindicatesignificantdif-
Earthsoils(openbars).Sixseedlingsperspecieswereusedper ferencesbetweenthetwosoils(t-test,p≤0.05)
concentrations (Ca ) and leaf N concen- significance(Adj-R2:0.56;p=0.052);butnorelationship
ADE-nonADE
tration (N )(Fig. 4); species showed either was found for Bnegative for P^ species (Adj-R2:-0.04;
ADE-nonADE
a higher (positive values on y axis, Fig. 4b) or a p=0.42) (Fig. 5b). Likewise, the relationship between
lowerleafNconcentrationinADEthaninnon-ADE leaf Ca concentration and RGRb showed that Bpositive
soils (negative values on y axis, Fig. 4b). The latter forN^speciesincreaseinRGRbwithincreasingleafCa
relationship, however, was only a trend, but not concentration(Adj-R2:0.86;p=0.01),butnorelationship
statistically significant (P=0.086, Fig. 4b). It is wasfoundforBNegativeforN^(Adj-R2:0.24;p=0.12)
worth to notice that most of the species located in (Fig. 5c). Furthermore, we found that there was not
the negative side of the y axis, where N-fixing significant relationship between leaf Ca concentration
legumes (Fig. 4a and b). We also found a positive and leaf N concentration neither in Bpositive for N^
relationshipbetweenthedifferenceinleafP andleaf species (Adj-R2: 0.47; p=0.12), nor BNegative for N^
N concentrations in ADE and non-ADE (Fig. 4c). species(Adj-R2:-0.05;p=0.46)(Fig.5d).
The relationship between leaf Ca concentration and
RGRb showed that Bpositive for P^ species increase in
RGRb with increasing leaf Ca concentration (Adj- Discussion
R2:0.64; p=0.034) (Fig. 5a), but no relationship was
found for Bnegative for P^ species, (Adj-R2:0.09; p= The aim ofthisstudy was toevaluatewhethertropical
0.26) (Fig. 5a). Besides, we found that in Bpositive for tree seedlings respond to the differences in nutrient
P^ species, leaf P concentration increased with leaf Ca concentrationsbetweenADEandnon-ADE.Seedlings
concentration, although this was at the edge of indeedadjustedtheirmorphologyandtissuechemistry
250 PlantSoil(2015)396:241–255
a
1 4 b
*AnadenantheraCedrCehlaoris*iaMOarcmhoaseirai*um Cedrela
PnonADE0 *SamSawnieetaenia HymenaCeaarinianSaapindus *Albizia NnonADE02 AnSawdieenteannitahera* Sapindus
Leaf PLeaf −ADE−2−1 *Enterolobium *Erythrina eaf NLeaf −ADE−2 HSyammeannEaCeenaaatr*eirnoilaonbaiumOM*rmacohsaiae*rium* *Albizia
L
y = 0.r221:0−. 206.18x −4 Chorisia *Erythrina y = 0.43r −2: 00..4197x
−3 p=0.041 p=0.086
0 2 4 6 8 0 2 4 6 8
Leaf CaADE−Leaf CanonADE Leaf CaADE−Leaf CanonADE
1 c
Chorisia *Machaerium *O*rmAnoasidaenanthera Cedrela
Sapindus
af PnonADE0 *Albizi*aSama*nEeanCtearroilnoibainuHamymenaSewaietenia
e
L1
− −
PADE
af
e2
L− *Erythrina
y = −0.06 + 0.21x
r2:0.39
3 p=0.014
−
−4 −2 0 2 4
Leaf NADE−Leaf NnonADE
Fig.4 Linearregressionsofthedifferencesinleafnutrientcon- againstdifferencesinleafNconcentration.Speciesthataremaked
centrationsbetweenAmazonianDarkEarthandnon-Amazonian with an asterisk (*) are N-fixing legumes reported in Table 1.
DarkEarths.(a)DifferencesinleafPagainstdifferencesinleaf Results of linear regression are provided. Concentrations units
Ca.(b)DifferencesinleafNconcentrationagainstdifferencesin aremgg−1
leaf Ca concentration; (c) Differences in leaf P concentration
inresponsetotheincreasedfertilityinADE,butthisdid (Jefferies and Willis 1964). This inhibition could be
notleadtohigherseedlinggrowthinADE.Thislackof causedbymetabolicdisordersrelatedtoenzymeinacti-
growth response seems to be determined by the inter- vationandtoadecreaseofPavailabilityformetabolism
playbetweentheuptakeofCa andtheuptakeofother due to the excess of cytosolic Ca (Grundon 1972;
essentialnutrientssuchasNandP. JefferiesandWillis1964;ZohlenandTyler2004).
Wepredictedthatinhigh-fertilityADE,plantswould
PlantresponsestoADE investinabovegroundlightcapture,whereasinthenon-
ADE,plantswouldinvestinbelowgroundnutrientcap-
In our study we found lower survival in seedlings of ture. This is in line with Brower’s hypothesis which
Poeppigia and Terminalia that were growing in ADE, statesthatplantsinvestincapturingtheresourcethatis
comparedtothosegrowingonnon-ADE.Seedlingsof inmostlimitingsupply(Brouwer1962).Wefoundthat
Poeppigia were not able to develop a root system on seedlings growing on ADE allocated more biomass to
ADEandtheseedlingsthatcouldestablishonnon-ADE leaves (LMF) than seedlings on non-ADE, but we did
showedsomechlorosis(E.Quintero,personalobserva- not findsupport for higher investmentin rootbiomass
tion). These characteristics can be associated with in- onnon-ADE.Additionally,wefoundhigherinvestment
abilityofplantstoregulateCauptakewhentheygrowin in leaf area per leaf mass (SLA) and per plant mass
soil with high Ca. Studies in calcifuges species have (LAR) on ADE. Greater biomass allocation to leaves
suggestedthatexcessofCainthetissuestranslatetoCa in combination with higher SLA and LAR on ADE
toxicity, that inhibits growth, and causes mortality (Table 3) could increase seedlings capacity to capture