ArticlePDF Available

The environmental triangle of the Cerrado domain: ecological controls driving shifts in tree species composition between savannas and forests



1) The Cerrado Domain of central Brazil houses the largest extent of savanna in the Neotropics, but despite its simple characterisation as a giant savanna, it contains considerable vegetation heterogeneity that is poorly understood. 2) We aimed to determine how vegetation types in the Cerrado diverge in their tree species composition and what role ecological factors play in driving compositional patterns. 3) We used a dataset of 1,165 tree species inventories spread across the Cerrado Domain, which come from six vegetation types that have a substantial arboreal component: woody savannas, dystrophic cerradão, mesotrophic cerradão, seasonally dry tropical forests, semideciduous forests and evergreen forests. We found three extremes in terms of tree species composition, with clear underlying ecological drivers, which leads us to propose a ternary model, the ‘Cerrado Vegetation Triangle’, to characterize woody vegetation in the Cerrado. At one extreme, we found that semideciduous and evergreen forests are indistinguishable floristically and are found in areas with high water availability. At another extreme lie seasonally dry tropical forests which are found on more fertile soils. At the third extreme, we found that all types of savanna, and dystrophic cerradão, are highly similar in tree species composition and are commonly found in areas of poor soils and high flammability. Mesotrophic cerradão is transitional in tree species composition between savannas and seasonally dry tropical forest. 4) The lack of variation in tree species composition attributed to climatic variables indicates that within homogeneous macroclimatic zones, many types of forest and savanna co-exist due to complex mosaics of local substrate heterogeneity and fire history. 5) Synthesis. Our findings highlight the complexity of forest-savanna transitions in the Cerrado Domain, with relevance for understanding the future of Cerrado vegetation under environmental change. If nitrogen deposition is extensive, some savannas may be more likely to transition to mesotrophic cerradão or even seasonally dry tropical forest whereas if water availability increases these same savannas may transition to semideciduous or evergreen forest. Our ‘Cerrado Vegetation Triangle’ model offers a simple conceptual tool to frame discussions of conservation and management.
Journal of Ecology. 20 1 8; 1–1 2 .   
© 2018 The Aut hors. J ournal of Ecolog y
© 2018 British Ecological Society
  Accepted:22Febru ary2018
DOI : 10.1111/1365-2745.1 2969
The environmental triangle of the Cerrado Domain: Ecological
factors driving shifts in tree species composition between
forests and savannas
Marcelo Leandro Bueno1,2 | Kyle G. Dexter3,4| R. Toby Pennington4,5|
Vanessa Pontara2| Danilo Mesquita Neves6| James Alexander Ratter4|
Ary Teixeira de Oliveira-Filho1
1Depar tamentodeBotânica,ProgramadePós-G radua çãoemBiologiaVegetal,Universid adeFede raldeMinasGerais–UFMG ,BeloHo rizonte,B razil;
2Labor atoryofEcolog yandEvolut ionofPlants,Universid adeFede raldeViçosa,V içosa,MG,Brazil;3SchoolofGeosciences,TheUniversityofEdinburgh,
Edinbur gh,UK;4RoyalBotanicGardenEdinburgh,Edinburgh,UK;5D epartmentofGeography,TheUnive rsityofExeter,Exeter,UKand6Departmentof
EcologyandEvolut ionar yBiology,UniversityofArizona,Tucs on,AZ,USA
MarceloL eandroBueno
HandlingEditor:Giseld aDurigan
1. TheCerradoDomainofcentralBrazilhousesthelargestextentofsavannainthe
2. WeaimedtodeterminehowvegetationtypesintheCerradodivergeintheirtree
3. Weused a dataset of 1,165tree speciesinventoriesspread acrosstheCerrado
Domain, whichcome from six vegetation types that have a substantial arboreal
component: woody savannas, dystrophiccerr adão,mesotrophic cerradão, sea-
sonally dr y tropical forest s, semideciduous forests an d evergreen forests. We
ecological drivers, which leads us to propose a ternary model, the Cerrado
Vegetation Triangle,tocharacterizewoodyvegetationin theCerrado. Atoneex-
lie seasonal ly dry tropi cal forest s which are found on mo re fertile soi ls. At the
thirdextreme,wefound that alltypes of savanna, and dystrophic cerradão, are
highly simila r in tree species comp osition and are commo nly found in areas of
4. Thelack ofvariationin treespecies composition attributed toclimatic variables
histor y.
Journal of Ecology
BUENO Et a l.
The main factors considered as determinants of world-wide sa-
vanna distribution, composition and structure are fire, herbivory,
climate, soilfertility andwater availability, the lat ter beingaprod-
uct ofclimatic and edaphic factors(Bueno etal.,2017;Cole, 1986;
Collinson, 1988; Dantas, Batalha, & Pausas, 2013; Eiten, 1972;
ingfactorvaries fromone regionto another,most studies suggest
that climatic andedaphic factors are most critical (Lehmannetal.,
2014).While climatehas a macro-scale effect (Hirota etal.,2011),
Themain extentof Neotropicalsavanna is largelyfound within
Brazil where it is often termed the cerrado (Ab’Saber, 2003;
Gottsberger & Silberbauer-Gottsberger, 2006; Ribeiro & Walter,
“Domains,”andtheregionof centralBrazilthatisdominatedbysa-
anddr yseasons,andfiresarecommoninthedryseason,hindering
including both fire-tolerant and fire-dependent plants (Coutinho,
1990,2006; Durigan&Ratter,2006; Eiten,1972,1978;Hoffmann
etal., 2009; Simon etal., 2009). Most savanna-inhabiting woody
species have thick, corkybark and/orsubterranean meristems (x y-
lopodia), whichprotectthem from high temperaturesand allowre-
However,this widely used CerradoDomain” label hidesthe com-
plexityofvegetationfoundinthisregion, which ishighlyheteroge-
wellasdifferentt ypes of forest(Ab’Saber,20 03;DRYFLOR,2016;
Oliveira-Filho,Pennington,Rotella, &Lavin,2013;Oliveira-Filho&
Ratter, 1995, 200 0; Ratte r,As kew,M ontgomer y,& Gi fford, 1977;
Ratter & D argie, 1992; Ratter, Richa rds, Argent, & Gi fford, 1973;
Within the CerradoDomain,the speciescompositionof woody
plants i s expected to cha nge along a fire gr adient; in areas w ith-
out fire, species associated with forest environments commonly
out-c ompete savann a species (Dan tas & Pausas , 2013; Hoffmann ,
Orthen, & Nascimento, 20 03; Lehmann etal., 2014;Pausas,2014;
Silva, Co rrêa, Doane , Pereira, & Hor wath, 2013; Silva , Hoffmann,
etal., 2013), an d savanna can eve ntually conver t to forest ( Abreu
etal., 2017). In the absence of fire, the levels ofmineral nutrients
and water ava ilability are im portant fa ctors in the dis tribution of
vegetatio n types (O liveira-Filho & R atter, 2002). M ost soils of th e
Cerrado Domain are dystrophic, with low pH and high levels of
exchangea ble alumini um (Furley & Rat ter,1988; R atter, Ribeiro, &
Bridgewa ter,1997). O f the chemic al element s in the Cerr ado soil,
(2000). This element, of ten toxic toplants,occurs at high concen-
mationsshowhighlevels ofaluminiumtolerance(Meira-Netoetal.,
2017;Neri etal., 2012). Incontrast,speciesoccurring onlyin areas
with higher levels of calcium and magnesium and lower levels of al-
uminium ar e charac teristic of s ome kinds of fo rest in the C errado
andsemideciduousforests(Oliveira-Filho &Ratter,2002;Oliveira-
tions, anotherforest type,mesotrophiccerradão,is believedto act
asan intermediate community,intermsofbothsoilproperties and
species composition (Bueno, Neves, Oliveira-Filho, Lehn,&Ratter,
andtemporarily waterlogged areas withintheCerradoarecovered
byevergreenandsemideciduous forestsormarshy“campos”(cam-
occur in th e higher and bett er drained areas ( Amorim & Bata lha,
5. Synthesis. Ourfindings highlight the complexity of forest–savannatransitions in
theCerradoDomain,withrelevanceforunderst andingthefutureofCerradoveg-
dry tro pical forest , whereas if water availa bility increas es these same s avannas
may transiti on to semideciduous o r evergreen forest . Our “Cerrado Vegetat ion
tion and management.
cerradosensustricto,dystrophiccerradão,fire,galler yforest,mesotrophiccerradão,
Journal of Ecology
BUENO Et a l.
2007;Furley&Ratter,1988;Oliveira-Filho& Ratter,2002;Ribeiro
The varia tion in the ecolog ical factor s described ab ove in the
Cerrado Domain and their ef fect on the floristic composition
of vegetation types has been studied primarily at small spatial
scales , mostly at in dividual site s (e.g. Ratter, 1992; Ratte r,A skew,
studiesuptothe entire CerradoDomainbutusingaqualitativeap-
CerradoDomain (e.g. Bridgewater,Ratter, & Ribeiro, 2004; Ratter,
Bridgewater,&Ribeiro, 2003,2006; Ratteretal.,1997)thesehave
habitat sandmostforestvegetationtypes.Inaddition,theydidnot
include formal analyses of how environmental factors and fire
sites. We also analyse how compositional variation in tree species
correlat es with 27 climat ic and edaphi c variable s. Based upo n the
resultsofthese analyses,wedevelop a conceptual model that de-
typesintheCerrado Domain.Ourresultsarekeytounderstanding
forest–savanna transitions under global environmental changes,
suchasnitrogen deposition and increasing temperatures,andhave
relevancefor anyseasonallydr yregioninthe tropicswheresavan-
nasandforest sco-occur(Hirota etal.,2011; Lehmannetal., 2014;
2.1 | Study area
inSouthAmerica,surpassed inareaonlybythe Amazon(Ab’Saber,
2003;Gottsberger & Silberbauer-Gottsberger,2006),and spreads
across central Brazil, comprising c. 1/4 of the country’s surface,
plus smal ler areas in nort h-weste rn Paraguay and e astern Bolivia
(Oliveir a-Fi lho & Ratter, 200 2; Figure1). The Ce rrado Doma in ex-
tends over 20 degrees of latitude and altitudes ranging from 100 m
inthePantanalwetlands (central-western Brazil) to1,50 0minthe
tablelands of the Central Brazilian Highlands (Ribeiro & Walter,
2008). There is moderate variation in mean annual temperature
acrosstheDomain,rangingfrom18 to28°C. Variation inmeanan-
nualprecipitationisrelatively high, rangingfrom80 0to 2,000mm,
with amarked dryseason duringtheaustralwinter (approximately
We classified the vegetation of individual sites following the cri-
the veget ation of easter n tropical a nd subtropic al South Am erica.
Thissystemisafur therdevelopmentofthewidelyacceptedInstituto
Brazilianvegetation(Veloso,Filho,&Lima, 1991;reissuedby IBGE,
Within the Cerra do Domain, we samp led six main vegetati on
didnotinclude veget ation types that largelylack trees (e.g.campo
sujoorcampolimpo, c.f.Ribeiro& Walter,2008). Wegrouped the
various vegetation formationsthatcan be termed savanna,that is,
wi thagr a ss yun d ers tor eya nds om efr e qu e ncy off ir e ,a son ev ege ta-
dystrophicsoils, whichis largelysynonymous with the cerrado sen-
tido restritocategoryofRibeiroandWalter(200 8).Cerradãoischar-
tree cover), withtrees reaching a height of8–12m,and we distin-
guished t wo vegetation types for cerradão: dystrophic cerradão
FIGURE1 GeographicdistributionoftheCerrado(IBGE,2012),
Rondônia(RO),SãoPaulo(SP),SantaC atarina(SC),Sergipe(SE),
Journal of Ecology
BUENO Et a l.
on poor so ils and mesotro phic cerrad ão on richer soi ls (Ribeiro &
lar,buthavedistincttreespecies composition(Araújo,Nascimento,
Lopes, Rodrigues,& Rat ter,2011).Cerradão candevelop from cer-
vegetationformations oftenshare many tree species (Abreu etal.,
2017). In contr ast to cerrado s ensu stricto, t here is general ly not
a continuo us grassy layer in ce rradão, alth ough grasses a re often
present (Aristida, Axonopis, Paspalum and Trachypogon, Ribeiro &
sideredas forests (Oliveira-Filho&Ratter,2002;Ribeiro& Walter,
2008) , although they a re shorter in s tature than th e other forest
types found in the Cerrado Domain . Deciduous fores ts or SDTF
occuronscatteredpatches offertilesoils(morefertilethaninme-
sotrophiccerradão)and are notable for experiencing littlefire and
housing a markedly differentset of plantlineagesfrom other veg-
etation t ypes in the Cerrado (e.g. C actaceae; Bu eno etal., 2013;
Pennington,Prado,&Pendr y,2000;Ratteretal.,1977;Ratteretal.,
Cerrado Domainareevergreenandsemideciduousforests, largely
synonymous with mata de galeria and mata ciliar in the terminol-
ogyofRibeiroandWalter(2008),which are found in more humid
areas, suchasalongrivercourses(i.e.galleryforestandsemidecid-
uous ripa rian fores t), or in transi tion zones wit h the moist fo rests
ofeither theAmazon or Atlantic Forests (Ribeiro & Walter,2008).
Evergreen a nd semideci duous fores ts tend to be ri cher in specie s
than the ot her vegetati on types in t he Cerrado D omain (Olivei ra-
Filho&Rat ter,1995,2000,2002;Ribeiro&Walter,2008).
2.2 | Dataset
We extrac ted the dataset from the NeoTropTree (NTT ) database
for geo-referen ced sites, ex tending fro m southern F lorida (U.S .A.)
and Mexico to Patagonia. The NT T currently holds 6,000 sites/
and unpublished literature (e.g. PhD theses), particularly floristic
surveys and forest inventories of individual sites. Sites were as-
signed vegetation formations based on the classification used by the
Filho (2015, 2017). Sites are restric ted to a circular area with a
10-km diameter. Where two or more vegetation formations co-
herbariaand taxonomic monographs were added to the checklists
whentheywerecollectedwithin a5-kmradiusof the originalNTT
andgeographicalcircumscriptions,as defined (inthepresentcase)
bytheteamofspecialistsresponsiblefortheonlineprojectFlora do
they were cited in published checklist s. It also excludes species-
poorchecklists, whichisanimportantfilter becauselowsampling/
Thedatasetextracted from NT Tconsistedof 1,165 checklists,
sensu stricto),64asdystrophiccerradão,299as evergreen forest,
76assemideciduousforests, 140asSDTFand153asmesotrophic
cerrad ão (Figure1). The fi nal species matr ix contained pr esence/
ab s e n ced a t afo r 3 , 07 2 tree s p e cie s , with a tot a l of148,7 18p r e senc e
ment alvariablesforallsites,derivedfrommultiplesource s(ata3 0-
arc second resolution or c.1km2neartheequator).Elevationatthe
(Hijmans , Cameron, Pa rra, Jones , & Jarvis, 20 05), includin g mean
month, mi nimum temper ature of the cold est month, te mperature
annualrange, mean annualprecipitation,precipitation of thewet-
test month, precipitation of the driest month and precipitation
were deri ved from WorldCli m layers by Zomer, Trabucco, Bos sio,
vanSt ra aten,andVe rc ho t(20 08 ).World Climmonthlyte mp er at ur es
andprecipitationwerealsointerpolatedtoobtainvalues for5-day
intervalsby applying sinusoidalfunctionscentred atday15 bythe
meanvaluefore achmonth.Thesefunctionsyieldedvalu esfordays
1,5,10,20,25and30,which,inadditionto themeanvalueatday
15 ,w ere use dto ge n era teWal ter sC lim ate Dia gram s( W alt er,1985) .
able s:du ra ti on(n um berofd ay s)a ndseve rit y(mm)of bo th th ew at er
deficit andwaterexcessperiods.Daysoffrostwereobtained from
griddeddataset sproducedbyJonesandHarris(2008).
Surface rockiness (% sur face), soil texture class (%volume of
sand), salinity class (ECe in dS/m) and percent base saturation, a
proxyforsoil fertility,wereobtainedfromtheHarmonizedWorld
Soil Database v1.2 (available at
database-v12/en/) and ranked afterwards by mid-class percent-
inotherwords,theuse ofclasseswasadoptedtoaddrobustness
to the data because of the high local soil heterogeneity that can
make raw fig ures unrealis tic. The soi l drainage clas ses were ob-
tained following EMBR APA’sprotocol(Santosetal., 2013),which
combinessoil type,texture anddepthwithlandforms,inorderto
charac terize water availa bility. The season ality index rep resents
thesumof percentofrainfallacrossbothdeficit and excess peri-
odsfromWalterclimatediagrams. Thisindexisrelatedto climate
Journal of Ecology
BUENO Et a l.
features, drought and the effects of flooding (albeit indirectly).
fa ceonGo ogleEar t h©imagesinfive100×100mareas,oneatthe
dataweretr an sform ed tora nkedmi d- classvalue sforea ch si tean d
wereusedasa proxy for fire return interval (i.e. fire frequency;
Hoffmann etal., 2012; Archibald, Lehmann, Gómez-Dansd, &
2.3 | Data analyses
Toanalyse the floristicconsistencyofthe vegetationtypes,we ap-
plied non-metric multidimensional scaling (NMDS)ofspeciescom-
asthefloristicdissimilaritymetric .Inordertoimproveinterpret abil-
ity,ellipses showing99% confidence levels were addedaroundthe
vegetationtypecentroids.Multi-response permutation procedures
(MRPP)andanalysisofsimilarities(ANOSIM) wereusedtotestthe
The environmental variables were fit a posteriori to the NMDS
Theseanalyseswereconducted using theveganpackage (Oksanen
there are s ubsets of spec ies with signif icant associa tion with one
or more veget ation types . In this analysis , an indicator valu e (IV)
is derive d, with highe r IV values re presentin g greater aff inity of a
given species towardsa certain vegetation type. Thisanalysiswas
perfo rmed using the st atistical pack age indicspecie s (De Caceres
& Legendre , 2009) in the r St atistica l Environment (R C ore Team,
Several of t he main vegetation t ypes in the Ce rrado Domain we re
consistently discriminated in the NMDS ordination, indicating dif-
ferentiationintheirtreespecies composition,whileotherswerenot,
sensu stricto, comprising various savanna formations, grouped to-
gether in one corner of compositional space and was floristically
FIGURE2 (a)Non-metric
forest: green and semideciduous forest:
the correlations between the most
salinit y,soilsalinity;Sand,soilcoarseness
(%sand);SoilFertility,soilfertilit y(%base
seasonality; Water availability
NMDSforveget ationtypesandblack
fitted surface values for flammability
fitted surface values for Water Availability
Journal of Ecology
BUENO Et a l.
distinctfromSDTF andevergreen/semideciduousforests,which fell
inopposite corners ofthecompositional space.Dystrophiccerradão
grouped withcerradosensu stricto, from whichitwasindistinguish-
grouped withevergreen forests,from which they were composition-
ally indistinguishable. Mesotrophic cerradão was intermediate in
compositionbetweenthecerrado sensustricto/dystrophiccerradão
The stress value inthe two-dimensional NMDS was 0.11, indi-
cating that two dimensionswere adequate to represent the varia-
tion,andbasedonthestressplot,theoverallconfiguration fitsthe
Analysis of similarities and MRPP analyses that distinguished
significantlyintreespeciescomposition(ANOSIM,R=.76,p < .001;
MRPP, A=0.18, p<.001). When we categorized sites into three
major floristic groups: savanna/cerradão (cerrado sensu stricto,
ous/evergreen(evergreenandsemideciduosforests),theR- value of
ANOSIMincreased (R=.93, p<.001), indicating that threegroups
provide a bettercategorization ofsites than six groups.TheMRPP
Furthermore, wefoundthatseveralecologicalvariablesarekey
namelysoildrainageclass(relatedtowateravailabilit y),grasscover-
Themainindicator speciesanalysis yieldedsubsets oftreespe-
cies that are significantly associated with each of the vegetation
forest a re also frequent in s emideciduous f orests and vic e versa,
demons trating the ir floristic s imilarity. The s ame holds for in dica-
tor specie s of cerrado sensu s tricto being fr equent in dystro phic
cerradãoandvice versa.Meanwhile, indicatorspeciesformesotro-
phic cerradão have relativelyhigh frequencies inSDTF,dystrophic
cerradãoandsavanna sensustricto,demonstrating the transitional
and semideciduous forest s are scarce to absent in other vegetation
tationt ypesintheCerradoDomain.
the Cerrado Domain can be categorizedinto three principalfloris-
cerradão, SDTF,andevergreen/semideciduous forest, thelatter of
astheAmazon and Atlantic Forests (Oliveira-Filho & Ratter,2000,
2002; Mirandaetal.,inpress).The resultsclearlydemonstratethe
importance ofedaphic factorsinfacilitatingthecoexistenceof flo-
ristically divergentgroups undersimilar climaticregimes(Figure2),
which is evident from the completespatial interdigit ation of these
floristicgroupswithintheCerradoDomain(see also Mirandaetal.,
inpress). At any point in spacewithin the Cerrado Domain, one is
likely to be able to findallthree of thesefloristicgroups relatively
In order to highlight the edaphic factors influencing the tree
schematic that we refer to as the Cerrado Vegetation Triangle( CV T; 
Figure3). T he circle around t he triangle rep resents the b road cli-
matic envelopeofthe Cerrado Domain, which is stronglyseasonal
major factors that determine tree species composition. The arrow-
heads at th e vertices of th e triangle den ote extreme v alues for a
given ecol ogical fac tor that give ris e to each major fl oristic gro up
ofvegetationtypes:high firefrequencygivescerradosensu stricto
andcerradão,highsoilfer tilitygivesSDTFandhighwateravailabil-
tween these vertices.
FIGURE3 ProposedCerrado Vegetation Triangle related to
treespeciescompositioniscolouredaccordingtovegetationt ypes
Journal of Ecology
BUENO Et a l.
the tropics (e.g.Bond & van Wilgen, 1996; Coutinho, 1990; Dantas
etal., 2013; Gillon, 1983; Gottsberger & Silberbauer-Gottsberger,
2006;Platt,Ellair,Huffman,Potts,& Beckage,2016; Silva& Batalha,
in the savanna flora (Lehmann etal., 2014; Pennington & Hughes,
2009). Indicator species of savanna(cerrado sensu stricto), such as
Kielmeyera coriaceaMart. & Zucc.,Palicourea rigidaKunth,Byrsonima
coccolobifolia Kunth,Davilla elliptica A.St.-Hil., Dalbergia miscolobium
Benth and Zeyheria montanaMart. are characterized by thick corky
barkandsubterraneanmeristemsthatprotectthem fromhigh tem-
Gottsberger, 2006). In addition, the occurrence of these species is
correlated with soils of low fertility and high aluminium levels and
In the absence of fire, existing trees in a savanna (cerrado
sensu st ricto) have incr eased grow th and sur vival whi le addition al
tree individuals recruit. Thus, above-ground woody biomass and
tree den sity increase, i n a process termed wo ody encroachme nt.
Woody encr oachment is oc curring in tr opical sava nnas across t he
globe(Moreira,200 0;SanJosé&Fariñas,1991;Stevensetal.,2016;
Woinarski, Risler, & Kean, 20 04). In the context of the Cerrado
Domain , the increasing si ze and density of t rees often lea ds to a
forest fo rmation term ed cerradão (D urigan & Rat ter,20 06, 2016;
Pinheiro, Azevedo, & Monteiro, 2010; Pinheiro & Durigan, 2009,
2012). Given t hat many of the tree i ndividuals in ce rradão der ive
directlyfromacerradosensustrictovegetation,thesimilarit yintree
surprising(Figure4). Ifcerradãodoesexperiencefire,itmayrevert
tocerrado sensu stricto (Durigan&Ratter,20 06).The grassesthat
forestvegetationformation.Incontrast,thehighwateravailabilit yin
(cerrado s ensu stric to) than the oth er forest t ypes in the C errado
theotherforest formationsisrepresentedintheCVT byincreas-
higher water availability (i.e.low soil drainage).These fac tors can
interac t, and it has l ong been hypo thesized that s avanna form a-
tions on lowe r fertilit y soils are in herently mo re fire-prone than
vegetation on fertile soils, because of the slow rates at which
trees es tablish and gr ow,w hich then allows fl ammable grass t o
persis t in the communit y (Kellman, 1984; Lehmann e tal., 2014;
Pausas , 2014; Silva, Co rrêa, etal., 2013; Silv a, Hoffmann, e tal.,
2013). Forest fo rmations in the C errado Dom ain suppress fl am-
(Hoffmann&Moreira, 2002;Hoffmannetal., 2009).However,in
mesotrophic cerradão. In dystrophic cerradão, low soil fertility
may potentia lly limit th e maximum am ount of tree bio mass such
regime,because nutrientsmaybecome increasingly limiting as a
tree approaches the fire-resistance threshold (Hoffmann etal.,
2012; Pellegrini, Franco, & Hoffmann, 2016; Pellegrini, St aver,
Hedin, Charles-Do minique, &Tourgee, 2016).In addition to set-
availabilityalsoinfluencesthedistributionoftreespecies byreg-
ulating their growth rates and ability to overcome biomass loss in
a fire (Hof fmann etal., 2012 ; Lehmann, Arc hibald, Hoff mann, &
Mesotro phic cerradão i s found on soils inter mediate betwe en
thepoordystrophicsoilsofthesavanna formationsanddystrophic
cerradão and the mineral-rich mesotrophic or eutrophic soils of
the floristic gradient was controlled mainly by soil fertility. It may be
thatundercontinued fire exclusion,mesotrophiccerradão,through
litter deposition and nutrient cycling, may develop sufficient soil
FIGURE4 Speciesturnoveramongsixvegetationtypes.The
Journal of Ecology
BUENO Et a l.
fertility to transition to SDTF.These transitions between savanna,
mesotrophiccerradãoandSDTFm ustalsobeconsid eredinthecon-
Domain,whichcouldencouragewoodyencroachment and conver-
sion of savanna to forest vegetation.
In contra st with the smo oth transitio n from savanna to SDT F,
evergreenand semideciduous vegetation types is abrupt, notonly
ciduous fo rests (A dejuwon & Ades ina, 1992; Felfili & Sil va Junior,
1992;Furley,1976).The evergreen and semideciduous forests are
transit ion to non-forest veget ation is usua lly sharp. Th e transitio n
islessperceptible physiognomicallywhen it occurswith SDTF,but
Withincreasingtreesize,theamount ofnutrientsrequiredby
be espec ially limite d by nutrient s (Pellegri ni, 2016). For exampl e,
evergreen and semideciduous forest s have higher water availabil-
ity and a re associated wit h higher soil nut rient levels, pr omoted
by the highe r presence of c layey soil (A ssis, Coelh o, da Pinheir o,
tion ofwater availability andsoilfertility may explainthe distinc-
tiveindicator species from evergreen andsemideciduous forests
(e.g.Cheiloclinium cognatum(Miers)A.C.Sm.,Maprounea guianensis
Aubl., Calophyllum brasilienseCambess. for evegreen forests and
Garcinia gardneriana(Planch .&Triana)Zapp i,Hieronyma alchorneoi-
desAllemão,Unonopsis guatterioides(A.DC.)R.E.Fr.forsemidecid-
Walter, 2008). D espite present ing a similar tree sp ecies compo-
sition, these vegetation types differ in soil drainage, being better
forests(Ribeiro&Walter,2008;Rodrigues,2009).These vegeta-
tio nt yp esalsodif ferinthele af-fl us hre gi meand int hest ru ct ur eof
Rodrigu es & Shepher d, 2009). Th e CVT sug gests a cl ear floris tic
distinction between evergreen and semideciduous forests and sa-
vannaformations,wherethecausalfactorof vegetationchangeis
water availa bility and th e consequent ab sence of fire. The e ver-
green and semideciduous forests are also clearly floristically diver-
Seasonally dry tropical forest and evergreenand semidecidu-
ousforestsrelateto the edge oftheC VT with highersoilfertilit y
and/orgreater water availability(Ribeiro & Walter, 2008; Scariot
&Sevilha,2005).Evergreenandsemideciduous forests are more
associated with watercourses and wetter soils, whereas SDTFs
generallyhave noassociationwithstreams,butwithfertile soil in
species forSDTFsuch as Ximenia americanaL., Aspidosperma pyr-
ifolium Mart., Trichilia hirta L. and Amburana cearensis (Allemão)
A.C. Sm. are characteristic of higher soil fertility (Ratter etal.,
semideciduous formations show higherIVs,suggestinghighspec-
ificity for environmental factors such as water availability and soil
fertility. While we have noted that the transitions between ever-
green an d semidecid uous fores ts and othe r vegetatio n types ar e
generally abrupt in space, should precipitation patterns change
dramatically in the Cerrado Domain underglobal climate change,
Ouranalysessuggest that,within one climaticzoneintheCerrado
ity and a co mplex mosaic o f vegetations t ypes, which f orm three
vegetation(cerrado sensu stricto and cerradão), dry forests in high
fertility soils with low water availability (STDF)andseasonal or ev-
ergreenforests wheresoil water availabilityis high(evergreen and
semidec iduous fores t). We suggest a Cer rado Vegetation Triangl e
abi lityincont rolli ngthevaria tionintreespeci escom posit ionofveg-
etationtypesintheCerradoDomain.Thesefactorsactasimpor tant
filter s at local spat ial scales to i nfluence tre e species com position
acrossthe entireCerradoDomain,drivingareaswith highfirefre-
quencyandpoor soilstowardssavanna formationsand separating
two distinct forest formations related to soil fertility (SDTF) and
Much previous work has focused on the dist ribution of sa-
2014; Staver etal., 2 011),bu t has treated fo rest as one vege ta-
tion type. There are infact many kindsof forestin the Cerrado
Domain. Transi tions betwee n savanna and each o f these forest
mental drivers. In order to understand future transitions between
tinguishing the environmental driversand the kinds offorestin-
volved will be essential.
Thank you to r eviewers for prov iding very con structiv e revisions.
Thisstudywasinpartialfulfilment oftheDoctoralrequirements of
MLBwhothanksCNPqforsupporting a12-monthstudy periodat
the Royal Bot anic Garden Ed inburgh (gr ant SWE-202096/2011-4)
and Postdoctoral scholarship in UFMG (151002/2014-2). MLB
thanks the Royal Botanic Garden Edinburgh for support during
the time this research was conducted. KGD was supported by a
Journal of Ecology
BUENO Et a l.
DMN acknowledge support of NERC grant NE/I027797/1, DMN
(grant NSF/DEB-1556651).DMNalso thanks theBrazilian govern-
ment funding agencyCAPESfor supporting a 12-month studype-
M.L. B. and A.T.d.O.-F. conceived t he ideas; A.T.d.O.-F., J.A .R. and
M.L. B. compile d the data; M. L.B., V.P.a nd K.G.D. de signed meth-
odology;M.L .B.,D.M.N.andV.P.analysedthedata;V.P.andD.M.N.
commente d on earlier vers ions of the manusc ript; M.L .B., K.G. D.,
R.T.P. andA .T.d.O.-F.ledthewritingof themanuscript. All authors
contributed critically to the drafts and gave final approval for
Data used in this studyisavailable at
Marcelo Leandro Bueno
Abreu,R.C.,Hoffmann,W.A.,Vasconcelos,H.L.,Pilon,N.A .,Ross atto,
D.R.,&Durigan ,G.(2017). Thebiodiversitycost ofcarbon seques-
trationintropical savanna. Science Advances, 3, e1701284. https://
Ab’Saber,A.N.(20 03).Os domínios de natureza no Brasil: Potencialidades
Adejuwon,J.O.,&Adesina,F.A.(1992).Thenatureanddynamic softhe
forest-sav anna bound ary in sou th-wester n Nigeria. In P. A. Furl ey,
J. Proct or,& J. A . Ratter (Ed s.), Nature and dynamics of the forest-
savanna boundaries(pp.331–352).London,UK:Chapman&Hall.
Amorim ,P.K., & Bat alha, M.A. (2007). Soil-vegetation relationships in
Nationa l Park (centra l Brazil). Acta Ecologica, 32, 319–327.ht tps://
A r a ú j o , G . M . , & H a r i d a s a n , M . (1 9 8 8 ) . A c o m p a r i s o no f t h e n u t r i t i o n a l s t a -
tusoftwofores tcom munitiesonme sotroph ican ddys trophicsoils
incentral Brazil. Communications in Soil Science and Plant Analysis,
19,1075–1089.https://doi.or g/10.1080/0010 3628 80 93679 96
J. A. (2011). Str ucture an d florist ics of the ar boreal com ponent of
a dystrophic Cer radão and comparison with other Cerradões in
CentralBrazil.Edinburgh Journal of Botany,68,401–418.https://doi.
org /10.1017/S0960428611000242
Archibald, S., Lehmann, C. E. R ., Gómez-Dansd, J. L ., & Bradstocke,
R. A. (2013). Defining pyromes and global syndromes of fire re-
gimes. Proceedings of the National Academy of Sciences of the United
States of America, 110, 6442–6447.
pna s.121146 6110
Assis, A . C. C ., Coelho , R. M., da Pi nheiro, E. S ., & Durig an, G. (2011).
Water availa bility dete rmines physi ognomic gra dient in an area o f
low-fertility soils under Cerrado vegetation. Plant Ecology, 212,
1135–1147.https://doi .or g/10.10 07/s11258- 010 -989 3-8
Bond,W.J.,&vanWilgen,B.W.(1996).Fire and plants.London:Chapman
Bridgewater, S., Ratter, J. A., & Ribeiro, J. F. (200 4). Biogeographic
patterns, b-diver sity and dominance in the cerrado biome of
Brazil. Biodiversity and Conservation, 13, 2295–2318. https://doi.
org/10.1023/B:BIOC.0000 047903.37608.4c
Bueno,M.L.,Neves,D.R. M.,Oliveira-Filho,A.T.,Lehn,C.R.,&Ratter,
J. A. (2013). A s tudy in an area of tr ansition betwe en seasonally
dry tr opical fores t and mesotroph ic cerradão, in Ma to Grosso do
Sul,southwesternBrazil.Edinburgh Journal of Botany, 70, 469–486. 0164
Bueno, M.L .,Pennington,R.T.,Dexter,K.G.,Kamino,L.H.Y.,Pontara,
V.,Neves,D.M.,…Oliveira-Filho,A.T.(2017).Effectsofquaternar y
climaticfluc tuationson the distributionofneotropical savannatree
Cole, M.M. (1986).The savannas: Biogeography and geobotany(p.438).
Collinson,A .S.(1988). Introduction to world vegetation(2nded.,p.325).
Coutinho,L.M.(1990).FireintheecologyoftheBraziliancerrado.InJ.G .
Goldammer(Ed.),Fi re in the tropi cal biota( p p . 8 2 –1 0 5 ). B e r l i n , G e r m a n y : 
Coutinho,L.M.(20 06).Oconceitodebioma.Acta Botanica Brasilica,20,
Dantas,V.L.,Batalha,M.A.,&Pausas,J.G.(2013).Firedrivesfunc tional
thresholds on the savanna- forest transition. Ecology,94,2454 –2463.
Dantas,V.L .,&Pausas,J.G.(2013).Thelank yandthecorky:Fire-escape
strategiesins avanna woodyspecies.Journal of Ecology,101,1265–
1272 .h tt ps:// .12118
groups ofsites: Indicesandstatistical inference. Ecology,90, 3566–
DRYFLOR.(2016).Plantdiversitypat ternsinneotropic aldryforestsand
Dufrêne, M., & Legendre, P.(1997).Speciesassemblages and indicator
specie s: The need for a f lexible asy mmetrica l approach. Ecological
Duriga n, G., & Ratter, J. A . (2006). Successi onal changes in cer radão
andcerrado/forest ecotonalvegetation inwesternSãoPauloState,
Brazi l, 1962-20 00 . Edinburgh Journal of Botany,6, 119–130. https://
Cerradoconservation.Journal of Applied Ecology,53,11–15.https://
doi .org/10.1111/1365 -266 4.12559
Eiten, G.(1972). The cerr ado vegetation of Brazil. Botanical Review,38,
201–341.htt ps:// 07/BF0 285915 8
Eiten, G . (1978). De limitation of the cerrado concept. Vegetation, 36,
169–178.https://doi.or g/10.1007/BF02342599
sociologyandcomparisonofcerrado andgaller yforestsatFazenda
Agua Li mpa, Fede ral Dist rict, B razil. In P. A. Furl ey,J. P roctor & J.
A.Ratter(Eds.),Nature and dynamics of the forest-savanna boundaries
Furley,P.A. (1976).Soil-slope-plant relationshipsin thenorthern Maya
mountains, Belize,CentralAmerica III – Variationinthe properties
of soil profiles. Journal of Biogeography, 3, 303–319. https://doi.
Furley, P. A. (1992). Edaphic cha nges at the fores t- savanna bound ary
withpar ticularreferencetotheneotropics.InP.A.Furley,J.Proctor,
Journal of Ecology
BUENO Et a l.
&J.A.Ratter(Eds.),Nature and dynamics of the forest-savanna bound-
Furley, P.,& R atter, J. A. (1988). Soi l resource s and plant co mmunitie s
of the centr al Brazilian ce rrado and th eir developmen t. Journal of
Biogeography,15,97–108.ht tps://
Gillon , D. (1983). The f ire problem i n tropical s avanna. In F. Bourli ére
(Ed.), Tropical savannas (Ecosystem of the World 13) (pp. 617–641).
Gottsberger,G.,&Silberbauer-Gottsberger,I.(2006).Life in the Cerrado,
a South American tropical seasonal ecosystem. Vol. 1. Origin, structure,
dynamics and plant use.Ulm,Germany:Ret aVerlag.
Haidar, R. F., Fagg, J. M. F., Pinto, J. R. R ., Dias, R. R. , Damasco, G.,
Silva, L . C. R., & Fag g, C. W. (2013). Flore stas est acionais e áre as
deecótono no estadodoTocantins, Br asil: Parâmetrose struturais,
classificação das fitofisionomias florestais e subsídios para con-
servação. Acta Amzônica, 43, 261–290.
Haridas an,M.(2000).Nutriçãomineral das plantasnativasdoCerrado.
Revista Brasileira de Fisiologia Vegetal,12,54–64.
Very high re solution inter polated climate s urfaces for gl obal land
areas. International Journal of Climatology,25,1965–1978.https://doi.
org /10.1002/(ISSN)1097-0088
Hi rot a ,M . ,H olm gre n ,M . ,Van Nes ,E. H., &S c he f f er,M .( 2011 ).G lob alr e-
silience oftropicalforestandsavannatocritic altransitions.Science,
Hoffmann,W.A .,Adasme, R.A .,Haridasan,M.,Carvalho,M. T.,Geiger,
tality, governs the dynamics of savanna-forest boundaries under
freque nt fire in centr al Brazil. Ecology, 90, 1326–1337. htt ps://doi.
org /10.189 0/08 -0741.1
Hoffm ann, W. A., Ge iger, E. L., Got sch, S. G ., Rossat to, D. R., Silv a, L.
C.R .,Lau,O. L., …Franco,A .C.(2012). Ecologicalthresholdsat the
savanna-forest boundar y: How plant tr aits,resourcesand fire gov-
ernthedist ributionoftropicalbiomes.Ecology Letters,15, 759–768.
https ://doi.or g/10 .1111/j.1461-024 8. 2012. 0178 9.x
Hoffmann,W.A .,& Moreira,A .G.(2002).Theroleoffireinpopulation
cerrados of Brazil: Ecology and natural history of a neotropical savanna
Hoffmann, W. A., Orthen, B., & Nascimento, P. K. V. (2003).
Comparative fire ecology of tropical savanna and for-
est trees. Functional Ecology, 17, 720–726. ht tps://doi.
org /10.1111/j .1365-243 5.20 03 .0 0796 .x
IBGE (Ins tituto Brasileiro de Geografia e Estatística). (2012). Manual
Técnico da Vegetação Brasileira. Série Manuais Técnicos em Geociências
Jones, P. D., & Harris, I. C. (2008). Climatic Research Unit (CRU)
time-series dat asets of variations in climate with var iations in
other phenomena. NCAS British Atmospheric Data Centre,
date of citation. Retrieved from
Kellman, M . (1984). Synergistic relationships between fire and low
soil fer tility in Neo tropical Sava nnas: A hypothes is. Oecologia, 16,
15 8 –1 6 0 .
S.,Hoffmann,W.A., …Bond,W.J.(2014).Savannavegetation-fire-
climate relationships differ among continents. Science, 343, 548–
Deciphering the distribution of the savanna biome. New Phytology,
191,197–209.ht tp s://doi.o rg /10.1111 /j.1469- 8137.2 011.03 689.x
Mccune, B., & Grace, J. B. (20 02). Analysis of ecological communities.
eranttreesintheCerrado.Acta Botanica Brasilica,31,11–18.htt ps:// /10.159 0/0102-3 3062016 abb 0380
Mirand a, P.L . S., Olivei ra-Filh o, A. T., Pennin gton, R . T.,N eves, D. M.,
South America. Global Ecology and Biogeography.
Mistr y,J. (1998).Fire inthe Cerrado (savannas)ofBr azil: An ecological
revi ew. Progress in Physical Geography, 22, 425–44 8. https://doi.
org /10.117 7/030 913339802 20 04 01
Moreir a, A. G. (2 000). Ef fects of f ire protec tion on savan na struc ture
inCentralBrazil.Journal of Biogeography,27,1021–1029.https://doi.
Neri,A.V.,Schaefer,C.E.G .R.,Silva,A. F.,Souza, A.L.,Ferreira Junior,
W.G ., & Meir a-Neto, J.A. A . (2012). The influence of soils on the
Cerradovegetation.Edinburgh Journal of Botany,69,1–27.htt ps ://do i.
org /10.1017/S0960428611000382
Neves ,D.R .M.,D exter,K .G. ,Pe nn in gton,R .T.,B ue no ,M .L .,&Olivei ra -
Filho, A. T.(2015).Environmental and historicalcontrols offloris tic
composition across the South American Dr y Diagonal. Journal of
Neves, D.M.,Dex ter,K. G., Pennington,R. T.,Valente,A.S. M.,Bueno,
itats in theAtlanticForest Domain ofSouth America .Diversity and
Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P. R.,
O’Hara,R .B.,…Wagner,H. (2016). Vegan: community ecology pack-
age. R package version 2.0–3.Retrievedfromhttp://CRAN.R-project.
Oliveira-Filho, A. T. (2015). Um Sistema de classificação f isionômi-
co-ecológica da vegetação Neotropical. In P. V. Eisenlohr, J. M.
Felfili,M. M. R. F.Melo,L.A .Andrade& J.A.A. Meira-Neto (Eds.),
Fitossociologia no Brasil: Métodos e estudos de casos(Vol.2,pp.452–
473).Viçosa:EditoraUF V.
Oliveira-Filho, A. T. (2017). NeoTropTree, Flora arbórea da Região
Neotropical: Um banco de dados envolvendo biogeografia, di-
versidade e conservação. Universidade Federal de Minas Gerais.
Oli veira-F il ho,A.T.,C ardo so,D.,Sc hr ire,B .D. ,Le wi s,G .P.,Pen ni ng ton,
R. T.,Brummer,T. J., … Lavin ,M. (2013). Stability structures trop-
ical woody plant diversit y more than seasonality: Insights into
the ecology of high legume-succulent-plant biodiversity. South
African Journal of Botany, 89, 42–57. h ttps:// /10.1016/
Oliveir a-Filho, A . T., Penning ton, R. T., Rotella , J., & Lavin, M. ( 2013).
Exploring evolutionarily meaningful vegetation definitions in the
R.P.Burslem&W.D.Simonsen(Eds.),Forests and global change(pp.
Edinburgh Journal of Botany, 52, 141–194. ht tps://doi. org/10.1017/
Oliveira-Filho,A.T.,&Ratter,J.A. (200 0).Padrõesflorís ticosdasmatas
ciliare s da região do cer rado e a evoluç ão das pais agens do Bras il
Centra l durante o Quater nário Tardio. In R. R. Ro drigues & H. F.
LeitãoFilho(Eds.).Matas ciliares: Conservação e recuperação(pp.73–
woody floraofthecerradobiome. In P.S. Oliveira, & R. J. Marquis
Journal of Ecology
BUENO Et a l.
(Eds.), The cerrados of Brazil (pp. 91–120).N ew York, NY: Columb ia
Pausas,J.G.(2014).Alternative fire-drivenvegetationst ates. Journal of
Vegetation Science,26,4–6.
Pellegrini,A .F.A. (2016).Nutrientlimitationin tropicalsavannasacross
multiple scales and mechanisms. Ecology, 97,313–324. http s://doi.
org /10.1890/15-0869.1
Pellegr ini, A. F. A., Fra nco, A. C. , & Hoffmann , W.A . (2016). Shift s in
funct ional trai ts elevate ri sk of fire-drive n tree dieba ck in tropic al
savanna and forest biomes. Global Change Biology,22, 1 235–1243.
https ://doi.or g/10 .1111/gc b.13110
Pellegrini, A. F.A ., Staver, A. C.,Hedin, L.O., Charles-Dominique,T.,&
Tourgee, A. (2016). Aridit y, not fire, favors nitrogen-fixing plants
across tropical savanna and forest biomes. Ecology,97,2177–2183.
Penning ton, R . T.,& Hughes,C. E. (2014).The remarkable congruence
of new and old world savanna origins. New Phytologist, 204, 4–6.
https ://doi.or g/10 .1111/nph.129 96
Penning ton, R. T., Prado, D. E., & Pendry, C. A. (20 00).
Neotropical seasonally dry forests and quaternary vegeta-
tion changes. Journal of Biogeography, 27, 261–273. https://doi.
Pinheiro, M. H. O., A zevedo, T. S., & Monteiro, R. (2010). Spatial-
temporal distribution of fire-protectedsavannaphysiognomiesin
Southeastern Brazil.Anais da Academia Brasileira de Ciências, 82,
Pinheiro,E.S.,& Durigan, G. (200 9).Dinâmica espaço-temporal(1962–
sudestedoBrasil.Revista Brasileira de Botânica,32,441–454.
Pinheiro,E. S., &Durigan,G. (2012). Diferenças florístic ase estruturais
entrefitofisionomiasdoCerradoemAssis,SP,Brasil.Revista Árvore,
Platt, W. J., Ellair, D. P., Huffman, J. M., Pott s, S. E., & Beckage, B.
(2016). Pyrogenic fuels produced by savanna trees can engineer
humid savannas. Ecological Monographs, 86, 352–372. https://doi.
org/10.10 02/ecm.1224
R Core Team. (2017). R: A language and environment for statistical com-
puting. Vienna, Austria: R Foundation for Statistical Computing.
Ratter,J. A .(1971).Somenotes on twotypesof cerradão occurring in
northeastern Mato Grosso. In M. G. Fer ri (Ed.), III Simpósio Sobre o
Ratter,J. A.(1992). Transitions between cerrado and forest vegetation
inBrazil.InP.A. Furley, J. Proctor,& J. A .Ratter(Eds.), Nature and
dynamics of forest-savanna boundaries (pp. 417429). London, UK:
Ratte r, J. A., A skew, G. P.,M ontgomery, R . F. , & Gifford, D. R . (1977).
Observações adicionaissobre o cerradão de solosmesotróficos no
Brasil central. InM. G. Ferri (Ed.), IV Simpósio sobre o Cerrado: Bases
para a Utilização Agropecuária(pp.3 03–3 16).Sã oP a ul o,B raz il :E DUS P.
Ratter,J. A.,Askew, G. P.,Montgomery, R. F., & Gifford, D. R. (1978a).
Obser vationsonforest sofsomemesotrophicsoilsin centralBrazil.
Revista Brasileira de Botânica,1,47–58.
Ratter,J. A.,Askew, G. P.,Montgomery,R. F.,& Gifford,D.R . (1978b).
Obser vations on the veget ation of northeastern Mato Grosso
II. Forest and Soils of the Rio Suiá-Miçú area. Proceedings of the
Royal Society of London, 203, 191–208. ht tp s:// /10.1098/
rspb.1978. 010 0
Ratter,J.A., Bridgewater,S.,&Ribeiro, J. F.(2003).Analysisof theflo-
of the woody vegetation of 376 areas. Edinburgh Journal of Botany,
Ratter, J. A., Bridgewater, S., & Ribeiro, F. (2006). Biodiversity pat-
terns of th e woody vegetati on of the Brazilia n Cerrado. In R . T.
Penning ton, G. P.Lewis ,& J. A. Rat ter (Eds.), Neotropical savannas
and seasonally dry forests: Plant diversity, biogeography and con-
org /10.1 201/CRC SYSAS SSPE
Ratter,J. A.,&Dargie, T.C. D. (1992).Ananalysisoft he floristic com-
positionof26 cerradoareasin Brazil.Edinburg Journal of Botany, 49,
235–250. 0001608
Ratte r,J. A ., Ribeiro , J. F.,& Br idgewater, S. (1997). The B razilian cer-
rado vegetation and threats to its biodiver sity. Annals of Botany,80,
223–230. 06/anbo.1997.0469
Ratter, J. A., Richards, P. W., Argent, G ., & Gifford, D. R. (1973).
Obser vationsonthevegetationof thenortheasternMatoGrossoI.
Thewoodyveget ationtypesoftheXavantina-Cachimboexpedition
area. Philosophical Transactions of the Royal Society B,26 6, 449–492.
InS.M.Sano,&S.P.Almeida(Eds.),Cerrado: Ecologia e flora(p p.151–
Rodrigu es, R. R. (20 09). Uma discuss ão nomenclat ural das forma ções
ciliares.InR .R .Rodrigues&H.F.Leitão-Filho (Eds.),Matas ciliares:
Conservação e recuperação(pp.101–107).SãoPaulo,Brazil:EDUSP/
Rodrigu es, R. R ., & Shepher d, G. J. (20 09). Fatores Cond icionantes d a
ciliares: Conservação e recuperação(pp. 101–107). São Paulo, Brazil:
Ruggiero,P.G.C.,Batalha,M. A.,Pivelo,V.R., &Meirele s,S.T.(2002).
Soil-vegetation relationships in cerrado (Brazilian savanna) and
semideciduousforest,SoutheasternBrazil.Plant Ecology,160,1–16.
htt ps:// 819219386
of a Trachypogon savanna pr otected for 2 5 years. Acta Oecologica,
Santos,H.G.,Jacomine,P.K.T.,Anjos,L.H.C.,Oliveira,V.A .,Lumbreras,
J.F.,Coelho,M.R.,…Oliveira,J.B.(2013).Sistema brasileiro de classi-
ficação de solos(3rded.).Brasília:Embrapa.
Sc ari ot ,A.,&S ev ilh a, A .C .(2 0 05 ). Bi odi ve r si dad e, est rut ura ec on ser v ã o
defl or est asest a ci on aisde ci dua isno Ce rra do.I nA . Sc a ri ot ,J.C. Sou sa -
Silva&J.M.Felfili(Eds.),Cerrado: Ecol ogia, Biodivers idade e Conser vação
Silva,I.A.,&Batalha ,M.A.(2010).Woodyplantspeciesco-occurrencein
Braziliansavannasunderdifferentfirefrequencies.Acta Oecologica,
36,85–91. 04
Silva, L . C. R., Co rrêa, R. S. , Doane, T., Pereira , E., & Horwat h, W. R.
(2013).Unpre cedentedcarbonac cum ulationinminedsoils:T hesyn-
C.,&Horwath,W.R .(2013).Cansavannasbecomeforest s?Acoupled
and Soil,373,829–842.
Simon, M . F., & Pe nnington, R . T. (2012). Evidenc e for adaptation to
fire regimes in the tropical savannas of the Brazilian Cerrado.
International Journal of Plant Sciences, 173, 711–723. https://doi.
org /10.108 6/665973
Simon, M . F.,G rether, R., Que iroz, L. P., Skema, C. , Penningto n, R. T.,
& Hughes , C. E. (2009). Recent assemb ly of the Cerrado, a neo-
tropical plant diversit y hotspot, by insitu evolution of adaptations
to fire. Proceedings of the National Academy of Sciences of the United
States of America, 106 , 20359–2036 4.
pnas.0903 410106
St av er,A .C. ,Ar chi ba l d, S., &L ev in, S.A .(2 011 ). The gl o ba le x te n ta ndde-
termina nts of savanna and fo rest as alternative biome s tates. Science,
Journal of Ecology
BUENO Et a l.
Stevens, N ., Lehmann, C . E. R., Mu rphy, B. P.,& Du rigan, G. (2 016).
Savannawoody encroachment is widespread acrossthree conti-
nents. Global Change Biology,23,235–244.https://doi .or g/10.1111/
gcb.13 40 9
Veloso, H. P., Filho,A. L.R. R.,& Lima, J.C. A.(1991).Classificação da
vegetação brasileira, adaptada a um sistema universal (p. 124). Rio de
Janeiro,R J:IBGE.
Walter, H. (1985). Vegetation of the earth and ecological systems of the
org /10.1007/978-3- 642-96 859-4
Woinarski,J.C . Z., Risler,J., & Kean, L .(2004).Responseof vegeta-
tionandvertebratefauna to23 yearsof fire exclusionin atropi-
cal Eucalyptus openforest,NorthernTerritory, Australia. Austral
Ecology,29, 156–176. ht tp s://doi.o rg /10.1111 /j.1442-9993.20 04 .
Zomer, R. J., Trabu cco, A., Boss io, D. A., van Str aaten, O., & Verchot ,
L. V.(2008). Climate change mitigation:Aspatial analysis ofglobal
reforestation. Agriculture Ecosystems and Environment, 126, 67–80.
Additional Supporting Information may be found online in the
How to cite this article:BuenoML ,DexterKG,Pennington
between forest s and savannas. J Ecol. 2018;00:1–12.
https://doi .org/10.1111/1365-2745.12969
... Notably, the cerradão can be distinguished into two compositionally distinct categories viz. dystrophic cerradão and mesotrophic cerradão, associated with differences in soil nutrient status (Furley and Ratter 1988;Neri et al. 2012;Bueno et al. 2018). This distinction is related to increasing soil Ca availability towards the mesotrophic cerradão, derived from weathered limestone and in situ modification of calcareous parent material (Furley and Ratter 1988;Neri et al. 2012). ...
... This distinction is related to increasing soil Ca availability towards the mesotrophic cerradão, derived from weathered limestone and in situ modification of calcareous parent material (Furley and Ratter 1988;Neri et al. 2012). Floristic dissimilarities between dystrophic woodlands and savanna are less evident based on soil nutrient status and other environmental properties such as fire dynamics or water availability represent important factors in this distinction (Bueno et al. 2018). It is interesting to note, however, that even within the same vegetation type, there is evidence for gradients in soil nutrient status influencing species composition. ...
... The shifts in Cerrado vegetation types with increasing soil nutrient availability is associated with an incremental shift in foliar nutrient content (Viani et al. 2014;. Cation rich soils select for deciduous plants with nutritious leaves with a short lifespan (Bueno et al. 2018). These changes in leaf nutrition are associated with an increase in leaf mass per unit area (LMA) (Viani et al. 2014;Abrahão et al. 2019;Tameirão et al. 2021). ...
Full-text available
Background The Cerrado of central Brazil—the world’s largest Neotropical savanna – is comprised of a mosaic of highly heterogeneous vegetation growing on an extremely diverse geologic and geomorphologic background. Geomorphic processes under stable tectonic and climatic conditions facilitated the development of diverse edaphic properties, which interact with disturbance events to form unique vegetation types. Scope In this review, we detail how the geophysical environment affects soil formation and evaluate the mechanisms through which edaphic conditions control vegetation structure, floristic diversity and functional diversity. Conclusion The influence of geomorphic processes on edaphic properties has a marked impact on the ecology and evolution of plant communities. Species exhibit morphological and physiological adaptations that optimise their successful establishment in particular soil conditions. Furthermore, fire disturbance alters these soil-vegetation associations further regulating the structural nature of these communities. Therefore, we propose an integrative view where edaphic, chemical and physical properties act as modulators of vegetation stands, and these conditions interact with the fire regime. The knowledge of plant edaphic niches, their functional traits related to resource acquisition and use, as well as the interaction of edaphic properties and disturbance regimes is paramount to research planning, conservation, and successful restoration of the full diversity of Cerrado vegetation types.
... Simon et al., 2009). Like other drylands, Caatinga dry forests and Cerrado savannahs exhibit high plant species turnover mainly driven by soil properties and aridity levels (Bueno et al., 2018;Silva & Souza, 2018). As the most densely settled Neotropical semiarid region, the Brazilian drylands have experienced a trajectory of intense deforestation, forest degradation and increased aridity, leading to biodiversity loss, desertification and vulnerability of human populations, which rely on forest products and other ecosystem services to enhance their livelihood conditions (Silva et al., 2017;Strassburg et al., 2017). ...
... The Cerrado region covers ~2 million km 2 or 23% of Brazil's territory plus small areas in Paraguay and Bolivia (Figure 1; Pennington et al., 2006). The region encompasses a highly heterogeneous environmental and vegetation macro-mosaic, from open-canopy savannahs (the predominant vegetation type) to closed-canopy seasonally dry forests (Bueno et al., 2018). The relief is mostly comprised of extensive plateaus and valleys, extending from lowlands (~100 m) to highlands of up to 1500 m. ...
... The relief is mostly comprised of extensive plateaus and valleys, extending from lowlands (~100 m) to highlands of up to 1500 m. The climate is also seasonal, but generally wetter compared to the Caatinga, with higher annual precipitation (800-2000 mm) and less prolonged droughts (Bueno et al., 2018; see Figure S1). Average annual temperatures range from 18 to 28°C following a south-north gradient. ...
Full-text available
Aim Locally abundant species are typically widespread, while locally scarce species are geographically restricted—the so‐called abundance‐occupancy relationships (AORs). AORs help explain the drivers of species rarity and community assembly, but little is known about how variation around such relationship is driven by species traits and niche‐based processes, particularly in tropical woody plants. We tested the hypothesis that AORs in tropical dryland woody plants are positive and mediated by niche and functional traits along environmental gradients. Location The Caatinga dry forest and Cerrado savannah, Brazil. Methods We aggregated abundance and occurrence data into grid‐cells representing local (10‐km) to landscape scales (50‐km). We calculated species mean relative abundance at occupied grid‐cells (local abundance) and the proportion of grid‐cells occupied (occupancy), and estimated their niche breadth and marginality along multivariate environmental gradients. Results AORs were positive but weak at different scales in both regions due to some locally abundant but geographically restricted species, with most species being both locally and geographically rare. Cross‐species variation in local abundance was largely unpredictable, but occupancy was strongly driven by niche and functional traits, with a prominent negative effect of niche marginality. Geographically restricted species were associated with rare habitats, such as wetter and less intensively used habitats. Large seeds and abiotic dispersal favoured occupancy in Caatinga at small and large spatial scales, respectively, whereas species with conservative leaves were more widespread across scales in Cerrado. Main conclusions Woody plants in dry tropical biotas exhibit weak AORs, a pattern likely related to low habitat availability and dispersal limitation. Caatinga and Cerrado biotas emerge as environmentally structured at multiple spatial scales, with several habitat‐specialist rare species bearing specific regenerative and resource‐use traits and relying on conditions threatened by climate change and land‐use intensification. Examining AORs through the lens of niche, functional traits and spatial scales enables mapping patterns and drivers of species commonness and rarity, enhancing understanding of species assembly and providing tools for biodiversity conservation.
... Like many savannas and grasslands globally, local-scale plant diversity can be very high: da Silva Menezes et al. (2018) found 56 species in one square meter of Pampa, and densities of 30-35 plant species per square meter are common in grassy ecosystems across southern Brazil. High species turnover (e.g., beta diversity) is another key element of grassy ecosystems: plant community composition can change dramatically along environmental gradients (e.g., Bueno et al., 2018;Andrino et al., 2020;Devecchi et al., 2020;de Souza et al., 2021;Amaral et al., 2022). Lineage turnover along geological time also seems to be high in some grassy ecosystems of Brazil, where some plant clades show some of the highest diversification rates in the world . ...
... Savannas are the dominant vegetation within the Cerrado, but also occur in the Amazon, Pantanal, Atlantic Forest and Caatinga. Within their climatic range, savannas predominantly occupy nutrient-poor soils (Furley and Ratter, 1988), mainly oxisols or quartz sand soils, while seasonal (tropical dry) forests tend to occur on richer limestone soils (Oliveira-Filho and Ratter, 2002;Bueno et al., 2018). Close to watercourses, where moisture levels in the soil remain higher during the dry season or are permanently waterlogged, savannas are either replaced by gallery forests or grassy wetlands. ...
... Interacting with climate and soil, fire is a key factor shaping mosaic savanna landscapes, in which savanna and forest ecosystems may cooccur, depending on the frequency and intensity of fires (Staver et al., 2011b;Abreu et al., 2017;Bueno et al., 2018). By limiting tree growth and preventing forest expansion, fire defines plant community composition and structure (Ribeiro et al., 2019). ...
In Brazil, the country with the highest plant species richness in the world, biodiverse savannas and grasslands – i.e., grassy ecosystems, which occupy 27% of the country – have historically been neglected in conservation and scientific treatments. Reasons for this neglect include misconceptions about the characteristics and dynamics of these ecosystems, as well as inconsistent or regionally restricted terminology that impeded a more adequate communication about Brazil's savannas and grasslands, both within the country and internationally. Toward improved communication and recognition of Brazil’s diversity of ecosystems, we present the key drivers that control the main types of grassy ecosystems across Brazil (including in regions of the country where forests dominate). In doing so, we synthesize the main features of each grassy ecosystem in terms of physiognomy and ecological dynamics (e.g., relationships with herbivores and fire). We propose a terminology both for major grassland regions and for regionally relevant vegetation physiognomies. We also discuss terms associated with human land management and restoration of grassy ecosystems. Finally, we suggest key research needs to advance our understanding of the ecology and conservation values of Brazil’s grassy ecosystems. We expect that a common and shared terminology and understanding, as proposed here, will stimulate more integrative research that will be fundamental to developing improved conservation and restoration strategies.
... To assess seed dispersal mode (SD, %), we classified the species with seeds dispersed by either wind (wind dispersal) or animals (animal dispersal) (Peres, 2016). We also classified the species as savanna or forest species based on their common occurrence in different vegetation types in the Cerrado biome (Bueno et al., 2018;Mendonça et al., 2008). To assess differences in functional composition between juvenile and adult components, we calculated the Community-Weighted Mean (CWM) of two important traits indicative of environmental drivers and thus different habitat types in the savannas: total bark thickness and maximum tree height (Dantas and Pausas, 2020). ...
... The classification of the species with seeds dispersed (SD) by animals (A) or wind (W), and the main vegetation type of occurrence (VT) of the species (forest -F or Cerrado sensu stricto -C, i.e., savanna) are given. The main vegetation type of occurrence was classified based on Bueno et al. (2018). The seed dispersal mode of each species was classified based on Silva-Junior et al. (2005). ...
... Then, the community should change towards the increasing in abundance of species present in the juvenile component. (Bueno et al., 2018). Due to our inclusion criteria (DBH < 5 cm and Height > 1.0 m), some species were only recorded in the juvenile component, such as Bauhinia rufa, Bauhinia sp., Duguetia furfuracea, Rudgea viburnoides, and Vernonia sp. ...
Woody plant encroachment (WPE) is a process that lead to the transformation of savanna environments into forests, and in the threatened Central Brazilian savanna (locally called Cerrado) it is a result of inadequate conservation policies. Here, we compared the floristic and functional attributes of the adult (trees with diameter at ground level ≥ 5cm) and juvenile (trees with diameter at ground level < 5cm) components in a Cerrado sensu stricto to assess changes in a plant formation under a process of woody encroachment. We found that the adult and juvenile components had a mean Jaccard similarity index of 19% and PERMANOVA analysis showed a separation of two clusters (species of the adult component and species of the juvenile component), indicating high species dissimilarity between both components. We also found a higher percentage of forest species, with lower bark thickness and dispersed by animals in the juvenile component compared to the adult component. Our results indicate that under a process of WPE, forest species less adapted to stressful conditions and fire can establish in the juvenile component and may reflect environmental changes as increasing shade, reduced fire and lower temperatures. Considering that Cerrado is becoming hot and drier, our results alert that WPE can make Brazilian savanna ecosystems more vulnerable to global climate changes, since it selects species less resistant to fire. Our sampling approach is useful to detect further encroachment in Cerrado throughout short-term plant inventories.
... The Cerrado presents notable physiographic variation (e.g., Sano et al. 2019) and an associated number of vegetation types. These include grasslands, wetlands, savannas, and seasonally dry and wet forests (Ribeiro & Walter 2008), the occurrence of which depends on ecological factors at the local scale, such as soil fertility, water availability and fire regime (Bueno et al. 2018). Because of this complex mosaic of vegetation types, the Cerrado is a savanna-dominated biome with the richest flora in the world (Klink & Machado 2005). ...
... Seasonally dry forests occur as scattered patches within the Cerrado, often associated with limestone outcrops and high-fertility soils, which stand in contrast to the acidic and nutrient-poor soils that predominate in the Cerrado region. As a consequence, central Brazil dry forests, which are mostly deciduous during the dry season, differ markedly in species composition compared to adjacent savannas and wet forests (Pennington et al. 2000, Bueno et al. 2018. The characteristic dry forest species Aspidosperma subincanum and Tabebuia roseoalba were among those highly collected species in the IFN Cerrado, which also recorded other typical dry forest representatives mentioned in the literature, such as Aspidosperma pyrifolium, Commiphora leptophloeos, Machaerium scleroxylon and Schinopsis brasiliensis (Scariot & Sevilha 2005, Pereira et al. 2011. ...
Full-text available
The National Forest Inventory (Inventário Florestal Nacional-IFN) is a large initiative that uses standardised methods to survey Brazilian forestry resources. One target of the IFN is the Cerrado, which contains one of the richest floras in the world. The aim of this study was to assess the contribution of the IFN to the knowledge of Cerrado woody flora. We analysed data from field-collected vouchers sampled by the IFN Cerrado. We restricted our analyses to IFN collections of native trees and shrubs, including palms, which were identified at the species level. Habitat of each collection was obtained by overlaying specimens’ geographic coordinates with land cover maps available in the Mapbiomas platform. Our final dataset comprised 28,602 specimens distributed in 2,779 sites (conglomerates) in Bahia, Distrito Federal, Goiás, Maranhão, Mato Grosso, Mato Grosso do Sul, Minas Gerais, Piauí, São Paulo and Tocantins. Collections were located in the following habitats: savannas (40.5%), forests (30.2%), anthropic areas (25.6%), grasslands (3.5%), and water (0.2%). We recorded 1,822 species belonging to 543 genera and 105 families, representing 34% of Cerrado woody species recorded on Flora do Brasil 2020. Fabaceae had the largest number of species, while Tapirira guianensis and Matayba guianensis were the most collected species. We highlight 60 potentially new records of occurrence for several states and 64 new records for the Cerrado, primarily in riparian forests where species from other biomes occur. In addition, 232 recorded species are Cerrado endemics, while 36 are cited in the CNCFlora’s red list as endangered. The systematic sampling carried out by the IFN enabled vegetation sampling in remote and poorly known areas, which expanded the geographic range of many woody species and contributed to the knowledge of plant diversity in the Cerrado.
... Therefore, future studies assessing more inselbergs within Amazonia are needed to clarify similarity patterns and matrix permeability in this biome. Other investigations of beta diversity patterns in arid habitats within South America have highlighted water availability as one of the most relevant environmental drivers of species turnover (Neves et al., 2015;Bueno et al., 2018;Silva and Souza, 2018). Nevertheless, these investigations attributed less relevance to precipitation, and none found water availability to be the only relevant climatic predictor shaping species distribution, potentially indicating that precipitation patterns pose a greater challenge for plant establishment in inselbergs than in other arid environments, which could lead to the high permeability of arid lowlands. ...
Full-text available
Inselbergs are azonal formations found scattered in different biomes globally. The first floristic list focusing on an inselberg in the Brazilian Amazon is presented here. We aimed to investigate floristic and phylogenetic connections among Neotropical inselbergs and analyze whether environmental variables act as a filter of plant lineages. We used a database compiled from 50 sites spanning three main Neotropical biomes (Amazon, 11 sites, Atlantic Forest, 14 sites, and Caatinga, 25 sites) comprising 2270 Angiosperm species. Our data highlight the vastly different inselberg flora found in each biome. The inselberg floras of the Atlantic Forest and Caatinga show closer phylogenetic ties than those seen in the other biome pairs. The phylogenetic lineages found in all three biomes are also strongly divergent, even within plant families. The dissimilarity between biomes suggests that distinct biogeographical histories might have unfolded even under comparable environmental filtering. Our data suggest that the inselberg flora is more related to the biome where it is located than to other factors, even when the microclimatic conditions in the outcrops differ strongly from those of the surrounding matrix. Relative to the other biomes, the flora of the Caatinga inselbergs has the highest level of species turnover. There is a possibility that plants colonized these rather distant inselbergs even when they were found under very different climatic conditions than those in the Amazonian and Atlantic Forest biomes. It is worth noting that none of the studied inselbergs found in the Caatinga biome is protected. In view of the uniqueness and drought-resilient lineages present in each group of inselbergs, along with their vulnerability to destruction or disturbance and their strong connection with water availability, we stress the need to protect this ecosystem not only conserve plants potentially useful for ecological restoration but also to preserve the balance of this ecosystem and its connections.
... Data on N emissions [155] and deposition, and their impacts on South American ecosystems are scarce [156] . Additionally, most studies and observations are short-term [156,157] . Here we used the Dinerstein et al. [149] ecoregions and biomes classification to focus on the impacts of N deposition on grassland cover without including those on tree and shrub layers. ...
Full-text available
Grasslands are globally-important ecosystems providing critical ecosystem services. The species composition and characteristics of grasslands vary considerably across the planet with a wide variety of different grasslands found. However, in many regions grasslands have been impacted by atmospheric nitrogen deposition originating from anthropogenic activities with effects on productivity, species composition and diversity widely reported. Impacts vary across grassland habitats but many show declines in species richness and increases in biomass production related to soil eutrophication and acidification. At a continental level, there is considerable variation in the research effort that has been put into understanding the impacts of nitrogen deposition. In Europe, North America and parts of Asia, although there are unanswered research questions, there is a good understanding of N deposition impacts in most grassland habitats. This is not the case in other regions with large knowledge gaps in some parts of the world. This paper reviews the impacts of N deposition on grasslands around the world, highlighting recent advances and areas where research is still needed.
... In turn, the humid tropical forests, such as the ones in Amazonthe biggest and most diverse tropical forest on Earthare basically fire resistant due to high present humidity in its interior (Ray et al., 2005;Ray et al., 2010). Between these two biogeographic domainsthe most extense of the Neotropical regionthere is a wide range of contact where savannas and forests coexist, with the fire exerting an important ecological role in the maintenance of these two alternative states (Staver et al., 2011;Oliveras and Malhi, 2016;Bueno et al., 2018). ...
The Cerrado-Amazon transition (CAT) is located between the two largest biogeographic domains of the Neotropical region, coinciding with the “Arc of Deforestation”, an agricultural frontier that expands towards the Amazon. Fire plays an important role in the landscape changes that have been occurring in this transition. Thus, the objectives of this study were (i) evaluate the accuracy of different algorithms/techniques in predicting fires in the CAT; (ii) investigate the season of highest fire probabilities in the CAT; and (iii) identify, among anthropic, climatic, topographic and soil variables, the main drivers of fires spatial distribution in the CAT. Fire occurrence data (active fire pixels) from the MODIS sensor were used as input data in species distribution models. The SDM allowed the construction of monthly fire probability models for the study area. The model's quality was assessed by R-squared, RMSE, AUC and TSS metrics. In addition, we assessed the weight of variables in building the models. To obtain the fire seasonality, we used the bootstrap technique to calculate the 95% confidence interval (CI) about the mean. As result, we found that all algorithms obtained satisfactory performance, as well as the ensemble model technique used, which obtained an average AUC of 0.812 ± 0.001. The algorithm with the best individual performance was Random Forest, with a mean AUC 0.851 ± 0.002. Similar patterns were found when considering TSS metrics. The predicted and observed fire occurrences showed strong association (R² = 0.81). The high fire probability area was higher during the dry season (15304.37 ± 41.65), showing a significant difference (p-value ≤ 0.05) from the rainy season. The variables that most explained the fire occurrences along the CAT were 'Silt Fraction', 'Distance from Agriculture' and 'Maximum Temperature'. In addition, ‘Distance from Roads’, ‘Distance from Protected Areas’ and Elevation’ were relevant for the monthly models. Anthropogenic variables were more important during the dry season. The results demonstrate that SDMs can be adequately used in fire prediction. Protected areas have shown an important role in acting as a fire barrier. However, many of these areas are vulnerable, surrounded by high fire probabilities during the dry season, predominantly anthropic, putting the natural dynamics of fire in the region and the local biodiversity at risk. Our approach can be useful in environmental protection and monitoring, in addition to facilitating the understanding about the CAT ecological complexity and the high anthropogenic pressure happening there.
... Gallery forests showed a twofold increase in NDVI variability after the fire (burned gallery forest prefire NDVI SD = 0.07, postfire NDVI SD = 0.148) and postfire NBR variability in gallery forest vegetation (SD = 0.19) was more than twice that observed for savanna vegetation (SD = 0.07). At the time of the fire, environmental heterogeneity, such as microclimate and fine fuel availability, which influence fire severity, may have been higher in gallery forests than in savanna vegetation (Lentile et al. 2007;Bueno et al. 2018;Parks et al. 2018). In addition, the continuity of fine fuel in savanna areas in the late dry season may have contributed to the less variable fire behaviour (Price et al. 2014;Schertzer et al. 2015). ...
Fire can change gallery forest vegetation structure, thereby altering nutrient fluxes between terrestrial and aquatic ecosystems. In Brazilian savannas, which are fire‐prone ecosystems, the fire regime is changing due to human activities such as converting native vegetation to farmland and urban areas. Uncontrolled wildfires in these savannas can reach gallery forests, which are more sensitive to fire impacts, leading to concerns about the effects of fire on gallery forest vegetation structure and freshwater ecosystems. We analysed the relationships between fire severity (the degree to which the fire has changed an area), percentage of burned area, variation in precipitation with gallery forest and savanna vegetation structure recovery and input of nutrients (NH4, NO2−, NO3− and PO43-) to streams for 16 months after a fire event in five watersheds associated with small streams. One year after the fire, vegetation recovery (NDVI) was lower in gallery forest areas than in savanna woodlands, despite the more severe fires in savannas. The short‐term effects of fire on gallery forest vegetation included increased nitrate concentrations in streams, which were also influenced by increased precipitation and the extent of the burned area. The nutrient inputs into the stream stabilized within 1 year. However, gallery forest vegetation did not fully recover at that time and may continue to alter the functioning of the aquatic ecosystem. Together, these results demonstrate the need for an integrated fire management plan that considers both gallery forests and the surrounding savannas in the landscape to address consequences to aquatic ecosystems. O fogo pode alterar a estrutura da vegetação de florestas ripárias, alterando assim os fluxos de nutrientes entre os ecossistemas terrestres e aquáticos. Nas savanas brasileiras, que são ecossistemas propensos ao fogo, o regime de fogo está mudando devido a atividades humanas tais como a conversão de vegetação nativa em áreas agrícolas e urbanas. Eventos de fogo descontrolado em savanas podem atingir as florestas ripárias, que são mais sensíveis aos impactos do fogo, levando a preocupações sobre os efeitos do fogo na estrutura da vegetação de tais florestas e nos ecossistemas aquáticos associados. Nós analisamos as relações entre a severidade do fogo (o grau em que o fogo alterou uma área), percentagem de área queimada, e a variação da precipitação com a recuperação da estrutura de florestas ripárias e da vegetação em áreas de savana, e a entrada de nutrientes (NH4, NO2, NO3−, e PO43−) nos cursos de água ao longo de 16 meses após um evento de passagem do fogo em cinco bacias hidrográficas associadas a pequenos cursos de água. Um ano após a passagem do fogo, a recuperação da vegetação (NDVI) foi menor nas áreas de florestas ripárias do que nas savanas, apesar do fogo mais intenso nas savanas. Os efeitos em curto prazo do fogo em florestas ripárias incluíram o aumento das concentrações de nitrato nos cursos de água, o que também foi influenciado pelo aumento da precipitação e pela extensão da área queimada. As entradas de nutrientes nos córregos estabilizaram no prazo de um ano. Contudo, a vegetação de florestas ripárias não havia se recuperado totalmente após esse período e podendo continuar a alterar o funcionamento dos ecossistemas aquáticos. Juntos, estes resultados demonstram a necessidade de um plano integrado de gestão de incêndios que considere tanto as formações florestais ripárias como as savanas circundantes na paisagem, a fim de abordar as consequências para os ecossistemas aquáticos.
Full-text available
Vegetation is a key biosphere component to supporting biodiversity on Earth, and its maintenance and proper functioning are essential to guarantee the well-being of humankind. From a broad perspective, a fundamental goal of vegetation ecology is to understand the roles of abiotic and biotic factors that affect vegetation structure, distribution, diversity, and functioning, considering the relevant spatial and temporal scales. In this contribution, we reflect on the difficulties and opportunities to accomplish this grand objective by reviewing recent advances in the main areas of vegetation ecology. We highlight theoretical and methodological challenges and point to alternatives to overcome them. Our hope is that this contribution will motivate the development of future research efforts that will strengthen the field of vegetation ecology. Ultimately, vegetation science will continue to provide a strong knowledge basis and multiple theoretical and technological tools to better face the current global environmental crisis and to address the urgent need to sustainably conserve the vegetation cover of our planet in the Anthropocene.
Full-text available
Tropical savannas have been increasingly viewed as an opportunity for carbon sequestration through fire suppression and afforestation, but insufficient attention has been given to the consequences for biodiversity. To evaluate the biodiversity costs of increasing carbon sequestration, we quantified changes in ecosystem carbon stocks and the associated changes in communities of plants and ants resulting from fire suppression in savannas of the Brazilian Cerrado, a global biodiversity hotspot. Fire suppression resulted in increased carbon stocks of 1.2 Mg ha⁻¹ year⁻¹ since 1986 but was associated with acute species loss. In sites fully encroached by forest, plant species richness declined by 27%, and ant richness declined by 35%. Richness of savanna specialists, the species most at risk of local extinction due to forest encroachment, declined by 67% for plants and 86% for ants. This loss highlights the important role of fire in maintaining biodiversity in tropical savannas, a role that is not reflected in current policies of fire suppression throughout the Brazilian Cerrado. In tropical grasslands and savannas throughout the tropics, carbon mitigation programs that promote forest cover cannot be assumed to provide net benefits for conservation.
Full-text available
Aim: We aimed to assess the contribution of marginal habitats to the tree species richness of the Mata Atlântica (Atlantic Forest) biodiversity hotspot. In addition, we aimed to determine which environmental factors drive the occurrence and distribution of these marginal habitats. Location: The whole extension of the South American Atlantic Forest Domain plus forest intrusions into the neighbouring Cerrado and Pampa Domains, which comprises rain forests (‘core’ habitat) and five marginal habitats, namely high elevation forests, rock outcrop dwarf-forests, riverine forests, semideciduous forests and restinga (coastal white-sand woodlands). Methods: We compiled a dataset containing 366,875 occurrence records of 4,431 tree species from 1,753 site-checklists, which were a priori classified into ten main vegetation types. We then performed ordination analyses of the species-by-site matrix to assess the floristic consistency of this classification. In order to assess the relative contribution of environmental predictors to the community turnover, we produced models using 26 climate and substrate-related variables as environmental predictors. Results: Ordination diagrams supported the floristic segregation of vegetation types, with those considered as marginal habitats placed at the extremes of ordination axes. These marginal habitats are associated with the harshest extremes of five limiting factors: temperature seasonality (high elevation and subtropical riverine forests), flammability (rock outcrop dwarf-forests), high salinity (restinga), water deficit severity (semideciduous forests) and waterlogged soils (tropical riverine forests). Importantly, 45% of all species endemic to the Atlantic Domain only occur in marginal habitats. Main conclusions: Our results showed the key role of the poorly protected marginal habitats in contributing to the high species richness of the Atlantic Domain. Various types of environmental harshness operate as environmental filters determining the distribution of the Atlantic Domain habitats. Our findings also stressed the importance of fire, a previously neglected environmental factor.
Introduction. Why and how do ecosystems burn? Surviving fires - vegetative and reproductive responses. Plant demography and fire I: Interval dependent effects. Plant demography and fire II: Event-dependent effects. Fire and the evolutionary ecology of plants. Fire, competition and the organization of communities. Fire and management. Fire and the ecology of a changing world. References. Index.