Plant functional and taxonomic diversity in European grasslands along climatic gradients

J Veg Sci. 2021;32:e13027.  
|
1 o 12
h ps://doi.o g/10.1111/j s.13027
Jou nal o Vege a ion Science
wileyonlinelib a y.com/jou nal/j s
Recei ed:10Sep embe 2020
|
Re ised:11Ap il2021
|
Accep ed:13Ap il2021
DOI: 10.1111/j s.13027
SPECIAL FEATURE: MACROECOLOGY OF VEGETATION
Plan unc ional and axonomic di e si y in Eu opean
g asslands along clima ic g adien s
Coline C. F. Boonman1 | Luca San ini2,3  | Bjo n J. M. Rob oek4 | Selwyn Hoeks1 |
S e en Kelde man1 | Jü gen Dengle 5,6,7  | A iel Be gamini8 | Idoia Biu un9 |
Ma ia Lau a Ca anza10  | B uno E. L. Ce abolini11  | Milan Chy ý12  | U e Jand 13,14  |
Ta iana Lysenko15,16  | Angela S anisci17  | I ina Ta a enko18  | Sol i a Rūsiņa19  |
Ma k A. J. Huijb eg s1
1Depa men o En i onmen al Science, Ins i u e o Wa e and We land Resea ch, Radboud Uni e si y, Nijmegen, The Ne he lands
2Depa men o Biology and Bio echnologies “Cha les Da win”, Sapienza Uni e si à di Roma, Rome, I aly
3Na ional Resea ch Council, Ins i u e o Resea ch on Te es ial Ecosys ems (CNR- IRET), Mon e o ondo, I aly
4Depa men o Aqua icEcologyandEn i onmen alBiology,Ins i u e o Wa e andWe landResea ch,RadboudUni e si y,Nijmegen,TheNe he lands
5Vege a ionEcologyG oup,Ins i u eo Na u alResou ceSciences(IUNR),Zu ichUni e si yo AppliedSciences(ZHAW),Wädenswil,Swi ze land
6Plan Ecology, Bay eu h Cen e o Ecology and En i onmen al Resea ch (BayCEER), Bay eu h, Ge many
7Ge manCen e o In eg a i eBiodi e si yResea ch(iDi )Halle-Jena-Leipzig,Leipzig,Ge many
8WSLSwissFede alResea chIns i u e,Bi mensdo ,Swi ze land
9Plan BiologyandEcology,Facul yo ScienceandTechnology,Uni e si yo  heBasqueCoun yUPV/EHU,Bilbao,Spain
10En ixLab,Uni e si yo Molise,Pesche(IS),I aly
11Depa men o Bio echnologiesandLi eSciences(DBSV),Uni e si yo Insub ia,Va ese,I aly
12Depa men o Bo anyandZoology,Facul yo Science,Masa ykUni e si y,B no,CzechRepublic
13Ins i u eo Biology,Ma inLu he Uni e si yHalle-Wi enbe g,Halle(Saale),Ge many
14Ge manCen e o In eg a i eBiodi e si yResea ch(iDi )Halle-JenaLeipzig,Leipzig,Ge many
15Gene alVege a ionLabo a o y,Koma o Bo anicalIns i u eRAS,Sain -Pe e sbu g,Russia
16Phy odi e si yP oblemsLabo a o y,Sama aFede alResea chCen e RAS,Ins i u eo  heEcologyo  heVolgaBasinRAS,Toglia i,Russia
17En ixLab,Uni e si yo Molise,Te moli,I aly
18Schoolo En i onmen ,Ea handEcosys emSciences,Facul yo Science,Technology,Enginee ingandMa hema ics,TheOpenUni e si y,Mil onKeynes,
UK
19Facul yo Geog aphyandEa hSciences,Uni e si yo La ia,Rīga,La ia
Thisisanopenaccessa icleunde  he e mso  heC ea i eCommonsA ibu ionLicense,whichpe mi suse,dis ibu ionand ep oduc ioninanymedium,
p o ided heo iginalwo kisp ope lyci ed.
©2021TheAu ho s.Jou nal o Vege a ion SciencepublishedbyJohnWiley&SonsL donbehal o In e na ionalAssocia ion o Vege a ionScience
Thisa icleisapa o  heSpecialFea u eMac oecologyo Vege a ion,edi edbyMeelisPä el,F ancescoMa iaSaba ini,NaiaMo ue a-Holme,Holge K e andJü genDengle .
Co espondence
ColineC.F.Boonman,Depa men o 
En i onmen al Science, Ins i u e o Wa e
and We land Resea ch, Radboud Uni e si y,
POBox9010,NL-6500GL,Nijmegen,The
Ne he lands.
Email: C.Boonman@science. u.nl
Abs ac
Aim: Eu opean g assland communi ies a e highly di e se, bu pa e ns and d i e s o
hei con inen al- scale di e si y emain elusi e. This s udy analyses axonomic and
unc ional ichness in Eu opean g asslands along con inen al- scale empe a u e and
p ecipi a ion g adien s.
Loca ion: Eu ope.
2 o 12
|
Jou nal o Vege a ion Science BOONMAN e Al.
1 | INTRODUCTION
G asslands a e among he mos di e se ecosys ems in Eu ope (Wilson
e al., 2012; Dengle e al., 2020). They ha e been ex ensi ely s ud-
ied o a long ime and wi h long- e m moni o ing schemes (Scholz,
1975; Willems, 1983; Tilman e al., 2006). Va ious assembly p o-
cesses ha e been pu o wa d ha may explain he o igin and main-
enance o Eu opean g assland di e si y, e.g., compe i i e hie a chy
and niche pa i ioning (Mo i e al., 2018). Ye , o e he las 50 yea s,
Eu opean g assland di e si y has seen a d ama ic dec ease, o en
being a ibu ed o inc eased nu ien a ailabili y (Wesche e al.,
2012), o e g azing (Dengle e al., 2020) o d ough (Ca mona e al.,
2012; Noguei a e al., 2018). Insigh in o pa e ns o plan di e si y
o e en i onmen al g adien s is needed o aid deepe unde s anding
o he e ec s o global change on biodi e si y (e.g., Mooney e al.,
2009;Ca dinalee al.,2012;Funke al.,2017)andmayalsoimp o e
ou unde s anding o he mechanisms ha unde lie communi y as-
sembly(MacA hu &Le ins,1967;McGille al.,2006).
In he con ex o mac oecology, di e si y pa e ns o ege a-
ion a e ypically discussed ia a ious il e ing mechanisms (e.g.,
Weihe e al.,2011).Fi s , hedispe sal il e de e mines heabil-
i y o a species o be p esen in a speci ic loca ion, and hence he
egionalplan speciespoolandplan  ai pool(Cado e&Tucke ,
2017). Second, en i onmen al condi ions ac as an addi ional il e
on plan communi ies, so ing hose ha ul ill local ( undamen al)
niche equi emen scons i u edbyphysiologicalcons ain s(Leibold
e al., 2004; Tingley e al., 2014). Unde he a o abili y hypo hesis,
he mo e ex eme o un a o able en i onmen al condi ions a e, he
mo eselec i een i onmen al il e sa e(Fische ,1960).Thissug-
ges s ha only plan s wi h ai alues well adap ed o he ex eme
Funding in o ma ion
CCFB,MAJHandSHwe esuppo edby
he ERC p ojec (CoG SIZE 647224), IB by
heBasqueGo e nmen (IT936-16),MCby
heCzechScienceFounda ion(19-28491X),
andSRby heUni e si yo La iag an 
(AAp2016/B041//Zd2016/AZ03).
Co-o dina ing Edi o :MeelisPä el
Me hods: We quan i ied unc ional and axonomic ichness o  55,748 ege a ion
plo s.Sixplan  ai s, ela ed o esou ceacquisi ionandconse a ion,we eanalysed
o desc ibe plan communi y unc ional composi ion. Using a null- model app oach we
de i ed unc ional ichness e ec sizes ha indica e highe o lowe di e si y han
expec ed gi en he axonomic ichness. We assessed he a ia ion in absolu e unc-
ional and axonomic ichness and in unc ional ichness e ec sizes along g adien s
o minimum empe a u e, empe a u e ange, annual p ecipi a ion, and p ecipi a ion
seasonali y using a mul iple gene al addi i e modelling app oach.
Resul s: Func ional and axonomic ichness was high a  in e media e minimum
empe a u es and wide empe a u e anges. Func ional and axonomic ichness
was low in co espondence wi h low minimum empe a u es o na ow empe a u e
anges.Func ional ichnessinc easedand axonomic ichnessdec easeda highe 
minimum empe a u es and wide annual empe a u e anges. Bo h unc ional and
axonomic ichness dec eased wi h inc easing p ecipi a ion seasonali y and showed a
small inc ease a in e media e annual p ecipi a ion. O e all, e ec sizes o unc ional
ichnesswe esmall.Howe e ,e ec sizesindica ed ai di e gencea ex emelylow
minimum empe a u es and a low annual p ecipi a ion wi h ex eme p ecipi a ion
seasonali y.
Conclusions: Func ionaland axonomic ichnesso Eu opeang asslandcommuni ies
a y conside ably o e empe a u e and p ecipi a ion g adien s. O e all, hey ollow
simila pa e ns o e he clima e g adien s, excep a high minimum empe a u es
and wide empe a u e anges, whe e unc ional ichness inc eases and axonomic
ichness dec eases. This con as ing pa e n may igge new ideas o s udies ha
a ge speci ichypo heses ocusedoncommuni yassemblyp ocesses.And hough
e ec sizes we e small, hey indica e ha i may be impo an o conside clima e
seasonali y in plan di e si y s udies.
KEYWORDS
en i onmen al il e ing, a ou abili y hypo hesis, unc ional ichness, g assland di e si y,
limi ing simila i y, null model, plan ai di e si y, p ecipi a ion g adien , seasonali y,
axonomic ichness, empe a u e g adien , ai - en i onmen ela ionship
|
3 o 12
Jou nal o Vege a ion Science
BOONMAN e Al.
condi ions pe sis , esul ing in a educ ion in communi y unc ional
di e si y in ex emecondi ions(de Belloe  al., 2009; May ield&
Le ine,2010;Shene al.,2016).A hi d il e maybebio icin e ac-
ions, ep esen ing a coun e - g adien wi h compe i i e in e speci ic
in e ac ions being he main d i e o plan communi y composi-
ion a low abio ic s ess ( he s ess g adien hypo hesis; Be ness
&Callaway,1994).While hiscompe i ionmaylead oexclusiono 
speciesand educed ai  a ia ion(G ime,2006;May ield&Le ine,
2010;Kuns le e al.,2012), acili a i ein e ac ionsha ebeensug-
ges ed odomina ewhenabio ics essishigh(Be ness&Callaway,
1994;B ooke &Callaghan,1998),leading ohighe  ai di e gence
andinc eased unc ionaldi e si y(Valien e-Banue &Ve dú,2007;
McIn i e&Faja do,2014).Finally, heoccu enceandabundanceo 
a plan may also be in luenced by empo al a ia ion in clima e o
o he s esso s(Díaz&Cabido,1997;González-Mo enoe al.,2015;
Fische e al.,2020).Plan scanoccu inaplaceaslongasspecies’
niche equi emen sa eme a some imeo  heyea ,i.e.,exploi ing
empo a ily emp y niches (Godoy e al., 2009). This expansion o he
ealized niche can be especially impo an o he di e si y o plan
communi iesin empe a e egions(Scheine &Rey-Benayas,1994;
B ei schwe d e al., 2018).
When desc ibing di e si y pa e ns, he ocus should no only
be on plan species iden i y (i.e., hei axonomy) bu also on plan
ai s.As ai sdesc ibeamo edi ec linkbe ween hepe o mance
o an o ganism, local en i onmen al condi ions, and a plan 's unc-
ioningin hecommuni y(Keddy,1992;Funke al.,2017), heyplay
a c i ical ole in de e mining local plan communi y di e si y (Tilman
e al., 1997; Weihe e al., 1998). Especially when non- andom p o-
cesses de e mine communi y assemblages, species di e si y is
no anadequa esu oga e o  unc ionaldi e si y(Díaz&Cabido,
2001).Howe e ,bo hspeciesand ai composi ion a yalongen-
i onmen alg adien s(e.g.,Raymundoe al.,2018),making hesep-
a a iono  axonomicand unc ionaldi e si ydi icul (e.g.,Hoope 
e  al., 2002; Pe chey & Gas on, 2002). To con ol o  po en ially
coinciden al simila i y in di e si y– en i onmen ends, speci ically
designed ield expe imen s a e needed o one can use null models
ha check o  unc ionaldi e si ypa e nsdi e en  om andom
samplingo species(Go elli&G a es,1996;Swensone  al.,2012).
Taken oge he , hene in luenceo  he a ious il e ingmecha-
nisms on axonomic and unc ional plan ichness along la ge en i-
onmen al s ess g adien s emains elusi e. Pa o his ambiguous
ques ionis he alueo conside ingplan  ai di e si yinaddi ion
o axonomicdi e si y.He e,weanalyzed unc ionaland axonomic
ichness in Eu opean g asslands along empe a u e and p ecipi a-
iong adien s.Wequan i ied heabsolu e unc ionaland axonomic
ichness di ec ly de i ed om he communi y da a, as well as he
e ec size o unc ional ichness. The e ec size indica es highe o
lowe unc ional ichness han expec ed gi en he obse ed axo-
nomic ichness(e.g.,Ha eye al.,1983;Gö zenbe ge e al.,2016).
Fi s ,we ocusedonannualminimum empe a u e,asex eme em-
pe a u es limi me abolic ac i i y, g ow h and cons ain lea size, and
husa elikely o il e  ai composi ion(Wen ,1953).Second,we
included annual p ecipi a ion, as d ough cons ains lea longe i y,
pho osyn he ice iciency,andseedmass(Sandele al.,2010).Las ,
we included annual empe a u e ange and p ecipi a ion seasonali y
o be able o assess he in luence o seasonali y on plan di e si y a
hecon inen alscale(Scheine &Rey-Benayas,1994;Godoye al.,
2009).
2 | METHODS
2.1 | Da a collec ion and cleaning
2.1.1 | Vege a ionplo s
Plan communi y da a (i.e., ege a ion- plo eco ds o ele és) we e
ob ained om he Eu opean Vege a ion A chi e (EVA; Chy ý
e al., 2016) on 2 Ma ch 2020. Rele és we e classi ied by he expe
sys em EUNIS- ESy o he habi a ypes o EUNIS (Eu opean Na u e
In o ma ionSys em)Habi a Classi ica ion(Chy ýe al.,2020).All
g assland habi a s (EUNIS g oup R) and coas al dune habi a s (EUNIS
habi a s N15, N16 and N17) we e selec ed. Plo s no classi ied o
anyo  hesehabi a  ypesbu assignedin heEVAda abase o he
phy osociological class Molinio- A hena he e ea (Mucina e al., 2016)
we e added o he da a se .
Da a co e ed he whole o Eu ope. The islands o Maca onesia
we e excluded o emo e any e ec s o isola ed oceanic islands
wi hsigni ican wi hin-islandspecia ionandendemism(Humph ies,
1979). We selec ed plo s eco ded be ween 1979 and 2013 o c e-
a e an op imal ma ch wi h clima ic da a. In addi ion, we selec ed
plo swi hageo e e encingunce ain ysmalle  han1km,which
u he imp o ed he ma ch wi h high- esolu ion clima ic da a. Such
adi ec linkbe weenplan communi yandclima eda a educes he
e ec so con ounding ac o slikehabi a he e ogenei y(Szilágyi&
Meszéna, 2009). We only included lowe ing plan s as he a ailable
ai s o his g oup a e consis en wi hin he da a se . By doing so,
we excluded e ns, b yophy es, lichens and ungi.
These selec ion c i e ia esul ed in a geo e e enced occu ence
da a se wi h p esences o all selec ed plan species in all selec ed
plo s.F om hisini ialda ase , we excludedplo swi hpa icula 
ecologicalcondi ions(Münkemülle e al.,2020)using he ollow-
ing c i e ia: (a) plo s in luenced by high salini y, iden i ied by he
p esenceo speciesknown o a o ex emesal ycondi ions(e.g.,
Suaeda ma i ima)we eexcluded,as hei composi ionislikelyd i en
by salini y a he han empe a u e o p ecipi a ion; and (b) plo s
wi h a co e o ee o all sh ub species, iden i ied as all plan spe-
cies wi h an a e age heigh o mo e han 3 m (ob ained om he
TRYda abase,seebelow), ha exceeded5%inco e .Allsou ceso 
ege a ion-plo da aa elis edinAppendixS1.
2.1.2 | T ai da a
Weob ained ai da a om heTRYda abase e sion5(Ka gee al.,
2020), con aining bo h public and es ic ed da a se s. We e ie ed
4 o 12
|
Jou nal o Vege a ion Science BOONMAN e Al.
oo  ai da a om heFine-Roo EcologyDa abase(FRED;I e sen
e al.,2018).Allda asou cesa elis edinAppendixS2.A e s and-
a diza ion o ai uni s, we me ged ai da a om he wo da a-
bases and calcula ed species a e ages. While a e aging excludes
in aspeci ic ai a ia ion and ai plas ici y, which de e mines a
la gepa o communi ydi e si y(Albe e al.,2010;Junge al.,
2010; Ross e al., 2017; Ba bou e al., 2019; Niu e al., 2020), his
s ep was necessa y as ai da a we e no a ailable o mos species
ac oss he di e en plo s.
Func ional ichnessdepends on heselec edse o  ai s. We
usedasequenceo selec ionc i e ialeading oou  inal ai da a
se .Fi s ,weselec ed ai sbasedon hei impo ance o  ege a-
i e g ow h and ep oduc ion (W igh e al., 2004; Díaz e al., 2016).
Then, we selec ed bo h abo e- and below- g ound ai s wi h good
da a co e age. Thi d, we emo ed ai s o he same plan o gan wi h
high co a iance (e.g., speci ic lea a ea and lea d y ma e con en )
ino de  oequalizeo ganand ai impo ancein he unc ionaldi-
e si ycalcula ion,keeping he ai wi h hehighes da aco e age.
Fou h,weselec ednume ic ai sonly ominimize hee ec so di-
mension choice o calcula e unc ional di e si y (Mai e e al., 2015).
Ou inal ai da a se con ains six plan ai s (Table 1).
As unc ional ichnesscalcula ions(seebelow)a esensi i e o
hecomple enesso  hespecieslis pe plo (Pakeman,2014)and
hus equi eacomple e ai da ase ,we illed hegaps(14.7%o 
o e allmissingda a)inou  ai da abase(Table1;AppendixS3).
Impu ing missing da a is conside ed a be e al e na i e o emo -
ing incomple e plo s, which can po en ially esul in sys ema ic bi-
ases(Nakagawa&F eckle on,2008).Weused hemice unc iono 
he micepackageinR-3.6.1,which eliesonbe ween- ai co ela-
ionsusingmul i- a ia eimpu a ionwi hchainedequa ions o ill
gaps. This me hod has been shown o ha e lowe impu a ion e o
and bias compa ed o o he mul iple impu a ion app oaches, espe-
cially when he pe cen age o missing da a is high (Penone e al.,
2014).Al houghgap- illing echniques equi eonlyone ai  alue
pe species, we only e ained species wi h a leas wo ai alues
o achie e a mo e accu a e impu a ion. This esul ed in he emo al
o comple e ele és i he ai alues o a leas one species in he
communi ydidno mee  his equi emen .This esul edin heda a
se ha ing 61,714 ele és con aining a o al o 2,884 species. P io
o he impu a ion, we log- ans o med all ai da a since some
app oaches could be sensi i e o da a wi h a ying scales in he
a iables (Penone e al., 2014). Then, we used he p edic i e mean
ma ching me hod o he impu a ion o all ai s o p ese e non-
linea ai – ai ela ionships. We an 24 mul iple impu a ions ( an
Buu en, 2012) whe e ai p edic ions we e upda ed i e imes in
hechainedequa ions oimp o e hequali yo  hep edic ions.
Because mul iple impu a ion pe o mance bene i s om he addi-
ion o o he ai s ha may be ela ed o hose o in e es , we also
included plan longe i y (ca ego ical ai ), as his ai was com-
ple e a he genus le el. The gap- illing s ep esul ed in 24 ai da a
se s, whe e he a ia ion be ween hem ep esen s he unce ain y
o  hegap- illing echnique.Quali a i elyconsis en  esul s om
di e en impu ed da a se s indica e ha he conclusions on he
di ec ionali y o he e ec s a e no in luenced by he gap- illing
p ocedu e.
2.1.3 | Clima eda a
We cons uc ed ou en i onmen al g adien s. The i s was daily
a e age minimum empe a u e (°C) o he coldes mon h o he yea
and hesecondwasannualp ecipi a ion(mm).Fo bo h,en i on-
men s we e assumed o be ha she a he low end o he g adien
(Wen , 1953). Thi d, we conside ed p ecipi a ion seasonali y (%),
de ined as he coe icien o a ia ion based on mon hly p ecipi a-
ion.Fou h,weconside edannual empe a u e ange(°C),de ined
as he absolu e di e ence be ween he daily a e age maximum
empe a u e o he wa mes mon h and he daily a e age minimum
TABLE 1 Lis o  unc ionalplan  ai sused.Missingda a o species ep esen  hepe cen ageo species ha ha enoda a o  ha  ai .
Missing da a o e all ep esen he pe cen age o gaps in he o e all da a se o ha ai
T ai Uni Desc ip iona Rela ed unc ion
Missing da a (%)
Species O e all
Speci ic lea a ea mm2 mg- 1 One- sided a ea o a esh lea di ided by i s
o en- d ied mass
Pho osyn he ic a e, d ough
ole ance
16.4 1.2
Lea ni ogen
concen a ion
mg g- 1 To al amoun o ni ogen pe uni o d y lea
mass
Pho osyn he ic a e, s ess
ole ance
46.9 13.3
Plan heigh m Sho es dis ance be ween he uppe
bounda y o he main pho osyn he ic
issues on a plan and he g ound le el
Ligh cap u e,abo e-g ound
compe i ion, dispe sal dis ance
3.0 0.7
Seed mass g O en- d ied mass o an a e age seed o a
species
Dispe sal dis ance, seedling
compe i ion
12.8 1.7
Speci ic oo leng h mm mg- 1 Ra io o oo leng h o d y mass o ine oo s Resou ceup ake,s ess ole ance 85.3 35.3
Roo ing dep h m Maximum soil dep h om which esou ces
canbeacqui ed
Resou ceup ake,d ough  ole ance 76.1 35.9
aS anda dizedde ini ion omPé ez-Ha guindeguye al.(2013).
|
5 o 12
Jou nal o Vege a ion Science
BOONMAN e Al.
empe a u e o he coldes mon h o he yea . O he clima ic a i-
ables we e also used in his s udy as addi ional in o ma ion (daily
a e age wa mes empe a u e o he wa mes mon h [°C], mean
empe a u ein hewa mes qua e o  heyea [°C],mean empe a-
u ein hecoldes qua e o  heyea [°C],annualmean empe a u e
[°C], empe a u eseasonali y).Allclima icda awe edownloaded
omCHELSA e sion1.2a a esolu iono 0.01deg ee(~1kmg id
cells;Ka ge e al.,2017).
2.2 | Di e si y
We used wo di e si y me ics, axonomic ichness and mul i- a ia e
unc ional ichness. Taxonomic ichness quan i ies he numbe  o 
species p esen in a communi y o sampling plo , while unc ional
ichness exp esses he minimum olume encompassing he mos
ex eme ai alues in an assemblage and is a commonly used me -
ic oquan i ychangesinassemblyp ocessesalongen i onmen al
g adien s(Mason&deBello,2013;K a e al.,2015;Münkemülle 
e al., 2020). These me ics sui he goals o desc ibing plan di e si y
along en i onmen al g adien s. To calcula e unc ional ichness, we
i s c ea ed a species- speci ic ai dis ance ma ix based on scaled
alues o he six selec ed ai s and using mul i- a ia e Euclidean
dis ances wi hou ai weigh ing. Second, we pe o med a P incipal
Coo dina eAnalysis(PCoA)on hedis ancema ix.Thi d,weused
he esul ingcoo dina eso  hePCoA obuildamul i- a ia e ai 
spec um o all ai s wi hin a plo . Each da a poin in he mul i-
a ia e ai spec um desc ibes he ai s o a species, and he dis-
ances be ween poin s ep esen he simila i y o di e si ica ion in
ai s be ween species. The global ai unc ional ichness was com-
pu ed as he smalles six- dimensional con ex hull enclosing all ai
aluesinanassemblage.Fo aspeci icplo , he unc ional ichnessis
quan i iedas he olumeo  hecon exhull ha enclosesallspecies
o ha indi idual plo (Villége e al., 2008; Mouillo e al., 2013).
In his unc ional ichness calcula ion me hod, selec ing he
numbe o axes esul ing om hePCoAwasa ade-o .Thiscalcu-
la ion needs a highe numbe o species pe plo han he numbe o
includedPCoAaxesand huscon exhulldimensions.Wechose o
include all six axes o a oid excluding any ai a ia ion. The emo al
o plo s wi h six o ewe species did no c ea e any bias as a low
FIGURE 1 Geog aphic dis ibu ion o ege a ion plo s used in his s udy. Colou s ep esen he minimum empe a u e (a) and annual
empe a u e ange (b) o each plo in °C, as well as he annual p ecipi a ion in mm (c) and p ecipi a ion seasonali y in pe cen ages (d). The
en i onmen aldis ibu iono allloca ionsisplo edinAppendixS4

6 o 12
|
Jou nal o Vege a ion Science BOONMAN e Al.
numbe o species pe plo occu ed along he en i e en i onmen al
g adien and ollowed he same pa e n as plo s wi h mo e han six
species(AppendixS4).Remo ing hesecommuni ies esul edin24
inal da a se s, each wi h 55,748 plo s and 2,830 di e en species
(Figu e1).
2.3 | Null model
We used a null model o es o po en ial e ec s o axonomic
ichness on unc ional ichness, e ealing non- andom pa e ns ha
indica e ecological p ocesses as opposed o andom pa e ns expec ed
bychance(Go elli&G a es,1996;Swensone  al.,2012;Chalmand ie 
e al.,2013).Hence,wecalcula eda unc ional ichnesse ec sizeusing
esul s om a null model which andomly shu les species ai alues
in he ai da ase bu  e ainsspecies’ ai combina ions.Speci ically,
hespecies’PCoAsco eswe e andomlyshu ledamongspecies(bu 
no wi hin species) om he en i e species pool (de Bello e al., 2012).
Thenumbe o speciespe communi ywaskep equal o henumbe 
o species obse ed o ensu e he assessmen o unc ional ichness
ends, independen o di e ences in axonomic ichness (simila ly o
Swensone  al.,2012;C a ene al.,2018).
The e ec sizes we e calcula ed using p obabili ies since he dis-
ibu iono  henullmodelwasno symme ic(AppendixS5;Ul ich&
Go elli,2010;Be na d-Ve die e al.,2012;Lho skye al.,2016).We
used hep obi link unc ion(Φ- 1) o he VGAMpackage:
whe e
He e, we es ima ed he one- ailed p obabili y o  he obse ed
unc ional ichness (obs) ha ing a lowe alue han expec ed com-
pa ed o he unc ional ichness alues esul ing om he null model
wi h1,000i e a ions(NULL)(Equa ion2).Whene ec sizecalcula-
ions esul in alues close o ze o (i.e., obs is close o he median o
he null dis ibu ion), unc ional di e si y is no di e en om he null
expec a iono no il e ingo plan  ai s.La ge de ia ionso e ec 
size alues om ze o indica e la ge di e ences be ween obse ed
and expec ed (by chance, i.e., wi hou any plan ai communi y
assembly p ocesses) unc ional ichness. Since we andomized ai
alues ac oss species, nega i e e ec size alues indica e less a i-
a ion in ai ange han would be expec ed based on he numbe o
species in he communi y, i.e., unc ional con e gence o unc ional
unde dispe sion (clus e ing). Posi i e e ec size alues indica e mo e
a ia ion in ai ange han would be expec ed based on he numbe
o species in he communi y, i.e., unc ional di e gence o unc ional
o e dispe sion (Be na d- Ve die e al., 2012; Swenson e al., 2012;
Lho skye al.,2016).
2.4 | S a is ical analyses
We used he a e age o he o iginal unc ional ichness alues and
e ec size alues in one g id cell (~1km2) i mo e han one plo was
sampled o educe possible e ec s o spa ial au oco ela ion. This
esul edin19,179da apoin sandg idcells(Figu e1).Fo eacho  he
24 impu ed da a se s, unc ional ichness, axonomic ichness, and
e ec sizes we e assessed along g adien s o minimum empe a u e,
empe a u e ange, annual p ecipi a ion and p ecipi a ion
seasonali y in mul iple gene alized addi i e models wi h a Gaussian
dis ibu ion using he gamm4 package. These models allowed o 
inding pa e ns in he da a wi hou a p io i hypo heses on he
shapeo  he ela ionship.Howe e ,we es ic ed hecu inesso 
he endsbyse ing hekno s(k) o 4 o p e en o e i ing. Since
he di e ence be ween he models was small (Appendix S6), we
a e aged he ai da a o e he 24 impu ed da a se s, ecalcula ed
unc ional ichness alues, e an he null model, and used hese da a
in ou plo s o unc ional ichness and e ec sizes so ha con idence
in e als could indica e model unce ain y ins ead o di e ences
be ween impu ed da a se s.
3 | RESULTS
The di e si y o g assland communi ies a ied conside ably o e
empe a u eandp ecipi a iong adien s(Figu e2;AppendixS7).We
ound unc ional and axonomic ichness o be lowe in a eas wi h
low minimum empe a u es and in a eas wi h na ow empe a u e
anges(Figu e2a,c).Theseco esponded oloca ionsinmoun ain
anges o wi h highe la i ude o longi ude and loca ions on islands
o  along he wes  coas  o  F ance, Belgium and he Ne he lands,
espec i ely(AppendixS9).Thissimila i yin endbe ween unc-
ional and axonomic ichness may be due o a subs an ial co ela-
ion ( =0.7;AppendixS8).Howe e ,a wideannual empe a u e
anges, unc ional ichness inc eased wi h inc easing minimum
empe a u es(Figu e2a),while axonomic ichnessdec easedwi h
inc easing minimum empe a u es a e eaching an op imum a a
minimum empe a u e o  a ound −5 °C (Figu e 2c). These plo s
we e loca ed in con inen al Eu ope, whe e he highes minimum
empe a u es (i.e., highes unc ional ichness alues) we e ound
along hecoas so sou he nEu ope(AppendixS9).No e ha  o 
all empe a u e a iables (annual maximum, maximum o he wa m-
es qua e ,annualminimum,minimumo  hecoldes qua e ,and
annual mean empe a u e), bo h ichness me ics showed he same
non-linea pa e nas o  heminimum empe a u e(AppendixS10).
Likewise, empe a u e a ia ion a iables(annual empe a u e ange
and empe a u e seasonali y) showed he same non- linea pa e n
(Appendix S10). Conside ing he a ia ion o  unc ional and axo-
nomic ichness ac oss p ecipi a ion g adien s, we ound a dec ease
in ichnesswi hinc easingp ecipi a ionseasonali y(Figu e2b,d;
Appendix S9). In addi ion, unc ional and axonomic ichness ap-
pea ed o ha e an op imum a a e age alues o annual p ecipi a ion,
(1)
ES
=Φ
−
1(p)
(2)
p
=
numbe (NULL <obs) +
numbe (NULL = obs)
2
1, 000
|
7 o 12
Jou nal o Vege a ion Science
BOONMAN e Al.
hough he a ia ion in unc ional ichness ac oss his g adien was
small(Figu e2b,d).O e all, empe a u ecouldexplainmo eo  he
a ia ion in unc ional and axonomic ichness han p ecipi a ion
(AppendixS11).
Assessing he a ia ionin ai di e si ywhilecon olling o 
he numbe o species in he assemblage, unc ional ichness e -
ec sizes showed an opposi e end om unc ional and axo-
nomic ichnessalongallen i onmen alg adien s(Figu e2e, ).
E ec size alues we e posi i e when minimum empe a u es
d oppedbelow−11°Co exceeded−1°C, ep esen ingsampled
g asslands in he Alps, Russia and Scandina ia, and in sou h-
wes e nEu ope espec i ely(AppendixS9).La ge e ec sizes
we e ound in loca ions wi h ex emely low minimum empe a-
u e and a low annual p ecipi a ion wi h ex eme p ecipi a ion
seasonali y. Sligh ly nega i e e ec size alues we e ound a
in e media e minimum empe a u es, whe e sampled g asslands
a emos lyloca edincen alEu ope(Figu e2e;AppendixS9).
In con as , e ec sizes did no show a clea pa e n along he
empe a u e angeg adien (Figu e2e),no did hey a ymuch
ac ossbo hp ecipi a iong adien s(Figu e2 ).
4 | DISCUSSION
The esul s o ou con inen al- scale s udy show ha Eu opean
g assland communi ies a y subs an ially in unc ional and axo-
nomic ichness along empe a u e and p ecipi a ion g adien s.
Highes  ichnessinbo h ai sandspecieswas oundincommuni-
ies expe iencing a o able, in e media e o wa m clima es. These
loca ions con ain mo e species wi h ai combina ions ha show
FIGURE 2 Hea mapsdepic ing i ed
esul s o h ee mul iple gene alized
addi i e models. Values o unc ional
ichness (a + b) indica e he loga i hm
o he o iginal alues, while alues o
axonomic ichness (c + d) and e ec sizes
o unc ional ichness (e + ) ep esen
he o iginal alues. In panels (a– d), ligh e
colou s indica e lowe unc ional ichness
aluesandda ke colou sindica ehighe 
unc ional o axonomic ichness alues.
In panels (e) and ( ), yellow o ed colou s
a e posi i e e ec sizes, and blue o
pu ple colou s a e nega i e e ec sizes.
Whi e spaces a e loca ions wi hou da a
and/o ep esen combina ions o clima ic
a iables ha do no exis anywhe e in
Eu ope
8 o 12
|
Jou nal o Vege a ion Science BOONMAN e Al.
g ea e unc ional a ia ion compa ed o o he loca ions. This
pa e n is expec ed unde he a o abili y hypo hesis o lowe
di e si y in ha sh (low minimum empe a u es and less p ecipi a-
ion) en i onmen s and highe di e si y in mo e benign (wa me
and mo e equen  p ecipi a ion) en i onmen s (Fische , 1960).
Addi ionally, hispa e nisinlinewi h axonomicand unc ional
di e si y pa e ns ound in non-Eu opean g asslands (Mo adi &
Oldeland,2019).No e ha Mo adi&Oldeland(2019) epo ed ha 
p ecipi a ion limi s plan di e si y mo e han minimum empe a-
u eind ya eas,whichcon as swi hou  esul s(AppendixS11).
This may sugges ha , when d ie a eas would ha e been included
in his s udy, axonomic ichness may dec ease mo e owa d mo e
wa e - limi ed si es and de e mine he di e si y pa e ns mo e han
empe a u e.Ano he explana iono  hepeakin unc ionaland
axonomic ichness a a e age minimum empe a u e alues is he
mid-domaine ec (Colwell&Hu ,1994;Colwell&Lees,2000).
He e, hegeome iclimi sin ela ion o hedis ibu iono spe-
cies inc eases he o e lap o species anges and hus o species
ichnessin hemiddleo  heg adien (Colwell&Lees,2000).We
assume his e ec o be o mino impo ance as g asslands ex-
end omsou hwes e nEu ope oMiddleandCen alAsia,many
g assland species ha e la ge geog aphical anges (ex ending be-
yond he a ea included in his analysis), and many included species
also occu in o he habi a s besides g asslands.
Rema kably, unc ional ichness alues u he  inc eased owa d
highe minimum empe a u es and wide empe a u e anges, while
axonomic ichness dec eased. En i onmen s wi h a wide empe a u e
ange may hold co- exis ing species wi h dissimila ai alues as hey
can exploi di e en empo al niches, he eby domina ing in di e -
en seasonswi hou ou compe ing heo he species(Figu e2e;Díaz
e al., 1999). This may explain he posi i e e ec sizes a high minimum
empe a u esandwide empe a u e anges(Scheine &Rey-Benayas,
1994;González-Mo enoe al.,2015).Ne e heless,suchahypo hesis
should be in es iga ed u he by inspec ing a ia ion in species abun-
dance in a communi y o e ime (Tilman, 1996; Pescado e al., 2015).
Fu he no e ha due o hedis ega do in aspeci ic a ia ionin his
s udy, we po en ially o e es ima e unc ional ichness in en i onmen-
ally cons ic ed a eas as adap i e ai s may be mo e simila be ween
species in hese loca ions due o en i onmen al il e ing unde he
a o abili y hypo hesis (G an  & Abbo , 1980; Swenson & Enquis ,
2009; Swenson e al., 2012). Con e sely, we may unde es ima e unc-
ional ichness in o he a eas as adap i e ai s may be mo e dissimila
be weenspeciesunde  hep incipleo limi ingsimila i y(Weihe &
Keddy,1995;S ubbs&Wilson,2004;Mason&Wilson,2006;Violle
e al.,2011).Fu u e ieldcampaignsmigh in es inlocal ai measu e-
men s o all species o enable he inclusion o in aspeci ic ai a ia-
ion(Albe e al.,2010;Niue al.,2020).
Func ional and axonomic ichness dec eased owa d highe 
minimum empe a u es and na ow empe a u e anges. Unde he
s ess g adien hypo hesis, he educed di e si y may indica e s ong
compe i iondue oa benign, wa mclima e(Be ness&Callaway,
1994).Howe e , empe a u e angeis known oa ec  unc ional
ichness(González-Mo enoe al.,2015)andmigh ac asas esso in
hese loca ions explaining he educ ion in unc ional and axonomic
di e si y unde he a o abili y hypo hesis. Since posi i e e ec
sizes we e ound in hese en i onmen s, i could be a mo e s ess-
ul en i onmen whe e acili a ion possibly causes ai di e gence.
This may ollow he same explana ion o he ai di e gence ound
a ex eme low minimum empe a u es. In addi ion, acili a ion was
also obse ed in an alpine s udy whe e g ea e a ia ion in empe a-
u e made o mo e s ess ul en i onmen s (Molenda e al., 2012).
Ne e heless, hese opposi e e ec s may sugges ha he loca ions
wi h high minimum empe a u es and na ow empe a u e anges
migh be es ic ed by addi ional s esso s ha we e no included in
hiss udy,likesumme  empe a u es,wind,sola  adia ion,o e en
di e en he bi o yle els.Expe imen sa e equi ed odisen angle
he causal mechanisms unde lying he unc ional and axonomic
ichness pa e ns. They may also de e mine i and which en i on-
men al ac o s de e mine he di e si y in hese Eu opean g asslands,
bu ou esul s indica e he impo ance o clima e seasonali y.
In e p e ing pa e ns in communi y di e si y along clima ic
g adien s is di icul , pa icula ly because he a ia ion in ou da a
explained by he empe a u e and p ecipi a ion a iables was ela-
i elylow(8–10%;AppendixS10).Resul sshouldbein e p e edwi h
cau ionandexpe imen sa e equi ed obackuppossibleexplana-
ions o di e si y pa e ns. The low explained a ia ion indica es ha
o he ac o s beside clima e a y be ween loca ions and play a la ge
ole in de e mining he plan di e si y o Eu opean g assland com-
muni ies, such as a ia ion in soil, he bi o y, managemen , landscape
his o y and a ious biogeog aphical in luences (e.g., Willems, 1983;
Bakke e al.,2006;Dainesee al.,2015).I shouldalsobeempha-
sized ha unc ional di e si y is always dependen on he ai s ha
a e included in he s udy, whe e esul s may change when di e en
ai s a e selec ed (Villége e al., 2008). This means ha he conclu-
sions ega ding unc ional ichness ends in his s udy depend on
hespeci icse o  ai sweselec ed.Fu u e esea chcouldbene i 
om hee e -g owinga ailabili yo  ai da a(Ka gee al.,2020)
and ege a ion- plo da a om o he ecosys ems (B uelheide e al.,
2019) o assess he gene ali y o plan communi y di e si y pa e ns
along en i onmen al g adien s and o assess how communi y assem-
bly may change in esponse o global wa ming (Mouillo e al., 2007;
Mason e al., 2011; Ca o d e al., 2020).
ACKNOWLEDGEMENTS
The s udy was suppo ed by he TRY ini ia i e on plan ai s (h p://
www. y-db.o g)and heEu opeanVege a ionA chi e(EVA;h p://
eu o eg.o g/e a- da abase). The TRY ini ia i e and da abase a e
hos ed, de eloped and main ained by Jens Ka ge and Ge ha d
Bönisch(MaxPlanckIns i u e o Biogeochemis y,Jena,Ge many).
TRY is cu en ly suppo ed by DIVERSITAS/Fu u e Ea h and he
Ge manCen e o In eg a i eBiodi e si yResea ch(iDi )Halle–
Jena–Leipzig. EVA is a da abase o  he Wo king G oup Eu opean
Vege a ionSu eyo  heIn e na ionalAssocia ion o Vege a ion
Science.We hankSe geiN.She eme 'e  o  hecon ibu iono The
GlobalLea T ai sda ase  oTRYandIlonaKnollo á o  echnical
assis ancewi h heEVAda a.
|
9 o 12
Jou nal o Vege a ion Science
BOONMAN e Al.
AUTHOR CONTRIBUTIONS
CCFB concei ed he esea ch idea, wi h signi ican  con ibu ions
o LS, BJMR and MAJH; CCFBga he ed da a andpe o med he
analyses. CCFB w o e he manusc ip  wi h majo  con ibu ions
omLS,BJMR,SH,MCandMAJH;allau ho scommen edon he
manusc ip .
DATA AVAILABILITY STATEMENT
The unc ional ichness da a and R sc ip s on he andomiza ion
and null model a e a ailable ia DANS-EASY. The o iginal ai 
da acanbe eques ed iaTRY.Vege a ion-plo da ase usedin
hisp ojec iss o edin heEu opeanVege a ionA chi eunde 
p ojec no. 97.
ORCID
Coline C. F. Boonman h ps://o cid.o g/0000-0003-2417-1579
Luca San ini h ps://o cid.o g/0000-0002-5418-3688
Bjo n J. M. Rob oek h ps://o cid.o g/0000-0002-6714-0652
Selwyn Hoeks h ps://o cid.o g/0000-0001-5619-3233
Jü gen Dengle h ps://o cid.o g/0000-0003-3221-660X
A iel Be gamini h ps://o cid.o g/0000-0001-8816-1420
Idoia Biu un h ps://o cid.o g/0000-0002-1454-0433
Ma ia Lau a Ca anza h ps://o cid.o g/0000-0001-5753-890X
B uno E. L. Ce abolini h ps://o cid.o g/0000-0002-3793-0733
Milan Chy ý h ps://o cid.o g/0000-0002-8122-3075
U e Jand h ps://o cid.o g/0000-0002-3177-3669
Ta iana Lysenko h ps://o cid.o g/0000-0001-6688-1590
Angela S anisci h ps://o cid.o g/0000-0002-5302-0932
I ina Ta a enko h ps://o cid.o g/0000-0001-6835-2465
Sol i a Rūsiņa h ps://o cid.o g/0000-0002-9580-4110
Ma k A. J. Huijb eg s h ps://o cid.o g/0000-0002-7037-680X
REFERENCES
Albe ,C.H.,Thuille ,W.,Yoccoz,N.G.,Soudan ,A.,Bouche ,F.,Saccone,
P. e al. (2010) In aspeci ic unc ional a iabili y: ex en , s uc u e
and sou ces o a ia ion. Jou nal o Ecology, 98, 604– 613. h ps://doi.
o g/10.1111/j.1365- 2745.2010.01651.x
Bakke , E.S., Ri chie, M.E., Ol , H., Milchunas, D.G. & Knops, J.M.H.
(2006) He bi o e impac  on g assland plan  di e si y depends on
habi a p oduc i i y and he bi o e size. Ecology Le e s, 9, 780– 788.
h ps://doi.o g/10.1111/j.1461- 0248.2006.00925.x
Ba bou ,M.A.,E landson,S.,Peay,K.,Locke,B.,Jules,E.S.&C u singe ,
G.M. (2019) T ai plas ici y is mo e impo an han gene ic a ia ion
in de e mining species ichness o associa ed communi ies. Jou nal
o Ecology, 107, 350– 360. h ps://doi.o g/10.1111/1365- 2745.13014
Be na d-Ve die ,M.,Na as,M.-L.,Vellend,M.,Violle,C.,Fayolle,A.&
Ga nie , E. (2012) Communi y assembly along a soil dep h g adi-
en : con as ing pa e ns o plan ai con e gence and di e gence
in a Medi e anean angeland. Jou nal o Ecology, 100, 1422– 1433.
h ps://doi.o g/10.1111/1365- 2745.12003
Be ness,M.D.&Callaway,R.(1994)Posi i ein e ac ionsincommuni-
ies. T ends in Ecology & E olu ion, 9, 191– 193.
B ei schwe d ,E.,Jand ,U.&B uelheide,H.(2018)Usingco-occu ence
in o ma ion and ai composi ion o unde s and indi idual plan
pe o mance in g assland communi ies. Scien i ic Repo s, 8, 9076.
h ps://doi.o g/10.1038/s4159 8- 018- 27017 - 9
B ooke ,R.W.&Callaghan,F.V.(1998)Thebalancebe weenposi i eand
nega i e in e ac ions and i s ela ionship o en i onmen al g adien s:
a model. Oikos, 81, 196– 201.
B uelheide,H.,Dengle ,J.,Jiménez-Al a o,B.,Pu schke,O.,Hennekens,
S.M.,Chy ý,M.e al.(2019)sPlo –Anew ool o global ege a-
ion analyses. Jou nal o Vege a ion Science, 30, 161– 186. h ps://doi.
o g/10.1111/j s.12710
Cado e,M.W.&Tucke ,C.M.(2017)Shoulden i onmen al il e ingbe
abandoned? T ends in Ecology & E olu ion, 32(6), 429– 437. h ps://doi.
o g/10.1016/j. ee.2017.03.004
Ca dinale,B.J.,Du y,E.,Gonzalez,A.,Hoope ,D.U.,Pe ings,C.,Venail,
P. e al. (2012) Biodi e si y loss and i s impac on humani y. Na u e,
486(7401), 59– 67. h ps://doi.o g/10.1038/na u e11148
Ca mona, C.P., Azcá a e, F.M., de Bello, F., Olle o, H.S., Lepš, J. &
Peco, B. (2012) Taxonomical and unc ional di e si y u no e in
Medi e anean g asslands: in e ac ions be ween g azing, habi a
ype and ain all. Jou nal o Applied Ecology, 49, 1084– 1093. h ps://
doi.o g/10.1111/j.1365- 2664.2012.02193.x
Ca o d,J.A.,Dwye ,J.M.,Palma,E.,Cowles,J.M.&Tilman,D.(2020)
Communi y di e si y ou weighs e ec o wa ming on plan coloniza-
ion. Global Change Biology, 26, 3079– 3090. h ps://doi.o g/10.1111/
gcb.15017
Chalmand ie , L., Münkemülle , T., Gallien, L., de Bello, F., Mazel, F.,
La e gne,S.e al.(2013)A amilyo nullmodels odis inguishbe-
ween en i onmen al il e ing and bio ic in e ac ions in unc ional di-
e si y pa e ns. Jou nal o Vege a ion Science, 24, 853– 864. h ps://
doi.o g/10.1111/j s.12031
Chy ý,M.,Hennekens,S.M.,Jiménez-Al a o,B.,Knollo á,I.,Dengle ,
J.,Jansen,F.e al.(2016)Eu opeanVege a ionA chi e(EVA):anin-
eg a ed da abase o Eu opean ege a ion plo s. Applied Vege a ion
Science, 19, 173– 180. h ps://doi.o g/10.1111/a sc.12191
Chy ý, M., Tichý, L., Hennekens, S.M., Knollo á, I., Janssen, J.A.M.,
Rodwell,J.S.e al.(2020)EUNISHabi a Classi ica ion:expe sys-
em, cha ac e is ic species combina ions and dis ibu ion maps o
Eu opean habi a s. Applied Vege a ion Science, 23, 648– 675. h ps://
doi.o g/10.1111/a sc.12519
Colwell, R.K. & Hu , G.C. (1994) Nonbiological g adien s in species
ichness and a spu ious Rapopo e ec . The Ame ican Na u alis ,
144(4), 570– 595.
Colwell, R.K. & Lees, D.C. (2000) The mid-domain e ec : geome ic
cons ain s on he geog aphy o species ichness. T ends in Ecology
& E olu ion, 15(2), 70– 76.
C a en,D.,Hall,J.S.,Be lyn,G.P.,Ash on,M.S.& anB eugel,M.(2018)
En i onmen al il e ing limi s unc ional di e si y du ing succession
in a seasonally we opical seconda y o es . Jou nal o Vege a ion
Science, 29, 511– 520. h ps://doi.o g/10.1111/j s.12632
Dainese,M.,Lepš,J.&deBello,F.(2015)Di e en e ec so ele a ion,
habi a agmen a ion and g azing managemen on he unc ional,
phylogene ic and axonomic s uc u e o moun ain g asslands.
Pe spec i es in Plan Ecology, E olu ion and Sys ema ics, 17, 44– 53.
h ps://doi.o g/10.1016/j.ppees.2014.09.002
deBello,F.,P ice,J.N.,Münkemülle ,T.,Lii a,J.,Zobel,M.,Thuille ,W.
e al.(2012)Func ionalspeciespool amewo k o es  o bio ice -
ec s on communi y assembly. Ecology, 93, 2263– 2273. h ps://doi.
o g/10.1890/11- 1394.1
deBello,F.,Thuille ,W.,Lepš,J.,Chole ,P.,Clémen ,J.-C.,Macek,P.e al.
(2009) Pa i ioning o unc ional di e si y e eals he scale and ex en
o ai con e gence and di e gence. Jou nal o Vege a ion Science,
20, 475– 486. h ps://doi.o g/10.1111/j.1654- 1103.2009.01042.x
Dengle ,J.,Biu un,I.,Boch,S.,Dembicz,I.&Tö ök,P.(2020)G asslands
o he Palaea c ic biogeog aphic ealm: in oduc ion and syn hesis.
In:Golds ein,M.I.,DellaSala,D.A.&DiPaolo,D.A.(Eds.)Encyclopedia
o he Wo ld’s Biomes. Volume 3: Fo es s – T ees o Li e. G asslands and
Sh ublands – Sea o Plan s.Ams e dam,NL:Else ie ,pp.617–637.