Available via license: CC BY 4.0
Content may be subject to copyright.
Ecology a nd Evolution . 201 8 ;1–1 6 .
|
1
www.ecolevol.org
Received:18May2017
|
Revised:3 0January2018
|
Accepted:9February2018
DOI:10.1002/ece3.3973
ORIGINAL RESEARCH
Structural and functional responses of plant communities
to climate change- mediated alterations in the hydrology of
riparian areas in temperate Europe
Annette Baattrup-Pedersen1 | Annemarie Garssen2 | Emma Göthe1,3 |
Carl Christian Hoffmann1 | Andrea Oddershede1 | Tenna Riis4 |
Peter M. van Bodegom5 | Søren E. Larsen1 | Merel Soons2
Thisisanop enaccessarticleundert hetermsoft heCreat iveCommonsAttr ibutionLicense ,whichpe rmitsuse,dist ributionandreproductioninanymedium,
provide dtheoriginalworkisproper lycited.
©2018TheAuthors.Ecology an d Evolutionpu blishedbyJohnWiley&SonsLtd.
1Depar tmentofBioscience,Aarhus
University,Silkeb org,Denmark
2Depar tmentofBiology,UtrechtUni versit y,
Utrecht,TheNetherlands
3SectionforEcologyand
Biodiversity,SwedishUniversityof
AgriculturalSciences,Uppsala,Sweden
4Depar tmentofBioscien ce,Aarhus
University,Aarhus,Denmark
5InstituteofEnvironmentalSciences,Leid en
University,Leide n,TheNetherlands
Correspondence
AnnetteBaattrup-Pedersen,Department
ofBioscience,AarhusUniversity,Silkeborg ,
Denmark.
Email:abp@bios.au.dk
Funding information
EuropeanUnion7thFr ameworkProjects
REFRESH,Grant/AwardNumb er:244121;
MARS,G rant/AwardNumber :603378
Abstract
The hydr o l og yof r iparia n areas c hange srapid l ythese y earsb e c a u seof clima t echa n ge-
mediated alterationsinprecipitationpatterns.In this study,weused alarge-scalein
situexperimentalapproachtoexploreeffectsofdroughtandfloodingonplanttaxo-
nomicdiversityandfunctionaltraitcompositioninriparianareasintemperateEurope.
Wefound significant effectsoffloodingand drought in all study areas, the effects
being most pronounced under flooded conditions. In near-stream areas, taxonomic
diversityinitiallydeclinedinresponsetobothdroughtandflooding(althoughnotsig-
nificantlysoinallyears)andremainedstableunderdroughtconditions,whereasthe
decline continued underflooded conditions. For most traits, we foundclear indica-
tionsthatthefunctionaldiversityalsodeclinedunderfloodedconditions,particularly
innear-streamareas,indicatingthatfewerstrategiessucceededunderfloodedcondi-
tions. Consistent changes in communitymean trait values were also identified, but
fewer than expected. This can have several, not mutually exclusive, explanations.
First,differentadaptivestrategiesmaycoexistina community.Second,intraspecific
variability was notconsideredfor any ofthe traits. Forexample, many species can
elongateshootsandpetiolesthatenablethemtosurviveshallow,prolongedflooding
butsuchabilitieswillnotbecapturedwhenapplyingmeantraitvalues.Third,weonly
followedthecommunitiesfor3years.Floodingexcludesspeciesintolerantoftheal-
teredhydrology,whereastheestablishmentofnewspeciesreliesontime-dependent
processes,forinstancethe dispersal and establishmentofspecies within the areas.
Weexpect thataltered precipitation patternswill have profound consequences for
riparian vegetationintemperateEurope.Riparianareaswillexperience lossoftaxo-
nomicandfunctionaldiversityand,overtime,possiblyalsoalterationsincommunity
traitresponsesthatmayhavecascadingeffectsonecosystemfunctioning.
KEY WORDS
climatechange,drought,flooding,lowland,plant,trait,vegetation
2
|
BAATTRUP- PEDERSEN E T Al.
1 | INTRODUCTION
Intemperateregions,suchasNor thernandCentralEurope,climate
change-associatedalterationsinprecipitationpatterns,withhigher
thanaverageprecipitationandlesssnow accumulationduringwin-
terandlowerthan averageprecipitationduringsummer,likelyme-
diate significant alterations in the hydrological characteristics of
lowland streams.Inwinterandearly spring,anincreaseinthefre-
quency,magnitu de,a nddurationofflowev entswillocc ur(Kar lsson,
Sonnenborg, Seaby, Jensen, & Refsgaard, 2015; van Roosmalen,
Sonnen borg, & Jensen , 2009; Thodse n etal., 2014), whereas the
freque ncyanddurationofdroug htperiodsareexp ec tedtoincrease
duringsummer(Andersenetal.,2006;Christensen&Christensen,
2007). Higher temperatures will likely intensify deficits in water
budgetsduringsummerthroughenhancedevaporationandevapo-
transpiration, both of which will intensif y water stress (Douville
etal.,20 02).Furthermore,highertemperaturesmayextendtheac-
tive growth period of plants as growth maystar t earlier in spring
and continue for a longer time,therebypossiblyexacerbating the
effects of flooding and droughts on natural ecosystems (Zwicke
etal.,2013).
Climatechangeef fectsonthestruc turalandfunctionalprop-
ertiesofriparianecosystemsremaintobemorefullyelucidated.
In cre asi ng awa re n essof theim por t a nceofwe tla ndsfo ranum b er
ofecosystemservices such as flood protection, water purifica-
tion, water availability via groundwater recharge, and biodiver-
sity has s purred new studi es into the funct ioning of wetlands
ina changingclimate (see Catfordetal., 2013;Kominoski etal.,
2013; Garsse n, Verhoeven, & So ons, 2014; Garssen, B aattrup-
Pedersen,Voesenek,Verhoeven,&Soons,2015foranoverview).
Most of the studies conducted so far investigate the effectsof
climatechangesonriparian community compositionwith focus
ontheresponseofasinglespeciesorrestrictedspeciesassem-
blages(Catfordetal.,2013;Garssenetal.,2014,2015).Arecent
extensive review of plant communit y responses showed that
prolonged flooding and increasedinundation depth of riparian
areas tri gger signifi cant shif ts in specie s composition t hat may
leadtoeither increased or decreased riparian species richness,
depending on the environmental characteristics of the areas
(Garsse n etal., 2015).In Gar ssen etal. (2015), sp ecies richnes s
was obser ved to genera lly decline at fl ooded sites in n utrient-
rich catchments and at sites previously exhibiting relatively
stable hydrographs (forinstance rain-fed lowland streams; see
e.g., Beltman, Willems, & Güsewell, 20 07; Baattrup-Pedersen,
Jensen,etal.,2013),whereasanincreaseinspeciesrichnesswas
detectedat floodedsitesin dryareas(e.g.,in deser ts andsemi-
arid climate regions where manys treams areintermit tent; see
e.g.,Stromberg,Hazelton, &White,2009;Horner,Cunningham,
Thomson, Baker, & Mac Nally, 2012). In contrast, almost all
studie s of the effect s of increased drou ght episodes on ri par-
ianplantcommunit yresponseshave showna declinein species
richness, particularly for herbaceous species (e.g., Stromberg,
Bagstad, Leenhouts, Lite,&Makings,2005; Westwood, Teeuw,
Wade, Holmes, & Guyard, 2006; reviewed in Garssen etal.,
2014).A>30-daydroughtperiodthreatensthesurvivalofmany
species a nd usually ent ails a strong re duction in ri parian plant
biomass,andahigh droughtintensity(i.e.,a3–4cmwatertable
decline per day)may impair riparian seedlingsur vival,thereby
producingrelatively rapid changes in riparian species composi-
tion(Garssenetal.,2014).
The functionaltrait characteristics of plant species will likely
determinewhetherthespeciesareabletosurviveunderchanged
environm ental condi tions (Cornwel l & Ackerly, 2009; Ju ng etal.,
2014).Hence, trait-based predictions of the response ofriparian
communities to climatechange arevaluable.In contrast to taxo-
nomicapproaches,trait-basedmethodsenablegeneralizations(i.e.,
identif ication of common re sponses) to be made a cross regions
(Catfordetal.,2013;Diazetal.,20 04).Awiderangeoftraits can
beusedtodescribetheresponsesofspeciestotheirenvironment,
anddifferenttraitsmaycapturedifferentaspect sofresourceuse,
habitat requirements, and stress responses (e.g., Suding etal.,
2006; Thuiller, Albert, Dubuis, Randin, & Guisan, 2010). Traits
related to li fe form charac teristics, gr owth forms, gr owth rates,
photosyntheticpathways,leafmorphology,andchemistryhaveall
beenusedtoidentifyplantresponsestoenvironmentalconditions
as they af fect speciesgrowth, survival, and reproductiveoutput
(deBello&Mudrak, 2013;Violle etal.,20 07;Westoby& Wright,
2006).
Inthisstudy,we exploredthe effects of an experimentally
alteredhydrologyonthetaxonomicandfunctionaltraitcharac-
ter is ti cs oft heveget at ionan ddeposite dseedsi nriparian areas.
To increase the predictive potential, we used a large-scale
experimental approach in which we manipulated water levels
to disentangle the ef fects of specific environmental changes
fromco-occurring environmentalcharacteristicsthatmay oth-
erwiseblur the responses (see Ackerly,2004;Douma, Bardin,
Bart holomeus, & Bode gom, 2012; Wright, Reich , & Westoby,
2003). Anadditionalstrengthofthisapproachwasthatthedi-
rectlarge-scalewaterlevel manipulationsappliedpermits cre-
ation of groundwater–surface water interactions resembling
those likely to occur in riparian areas under current and ex-
pectedratesofclimaticchange.Toidentifycross-regionalcon-
sistentpatternsresponses inthe vegetation,theexperimental
siteswerelocatedinDenmark, Germany,andtheNetherlands.
Insomepart softhesites,weexperimentallyincre asedf looding
inthewinter/springandinotherpartsofthesitesweincreased
droughtsinsummer.
We analyzed r egenerative t raits and veget ative trait s that
we expected would change under altered hydrological con-
ditions (F igure1). The selec tion of traits w as based on theo-
retical considerations: Hydrological alterations are likely to
affec t traits asso ciated with the a bility to incre ase the water
uptake and/orconservewateras well astraitsassociated with
the ability to survive conditions with water surplus (Douma
etal.,2012;Hough-Snee etal., 2015). Thevegetativetraitsin-
cluded leaf trait s(specific leaf area,size, and mass),roottraits
|
3
BAATTR UP- PEDERSEN E T Al.
(rooting depth and porosity), and canopy (maximum height)
that may show a n adaptive respo nse to cope with an al tered
hydrology. Under drought conditions, we expected that the
abun da nc eofspe ci es wi th ex tensiverootingdepthsandspec ie s
withdense stems,smalland thick leaves,andlowspecificleaf
areas woul d increase in ab undance. The se traits c an serve to
maximizewateruptakeandatthesametime reduce waterloss
asthe rate of transpiration generally decreases withdeclining
specific leaf area and leafmass (Wright etal.,2005; Swenson
& Enquist , 2007; Poorter & Mar kesteijn, 200 8; Douma etal.,
2012; Figure1). Under flooded conditions, we expecte d that
the abundance of species withtraits associated with the abil-
ity to lower the metabolic activity (the “quiescence strategy”)
oravoidunfavorableconditions (the “escapestrategy”;Bailey-
Serres & Voesenek,20 08)would increase. Therefore, we an-
ticipatedthat the abundance of tall specieswould increase as
thesehavemoreeasyaccesstoatmosphericoxygenthanshor t
species. Additionally, we expected that species able to form
porous roots or aerenchyma inadventitious rootstofacilitate
oxygentransporttotheapicalrootzone(Armstrong,Brandle,&
Jackson,1994)wouldincreaseinabundance,asthesetraitscan
be critic ally impor tant to mai ntain the exch ange of gas unde r
flooded conditions(Bailey-Serres & Voesenek,2008; Garssen
etal.,2015).Wealsoconsideredregenerativetraits associated
with theabilityto disperse under droughtand flooded condi-
tions,respectively,includingseedmass,volume,andbuoyancy.
Specifically, we expected that species witha high seed mass
would dec line in abundan ce with enhanced f looding conco m-
itantl y with an incre ase in specie s with a high se ed buoyanc y
an dvol ume,ref lec tingt he adapt ivev alu eofp ro duc in gl owmass
buthighvolumebuoyantseedsthat candisperseefficientlyby
water(Doumaetal.,2012).
Thespecifichypothesestestedwerethatfloodinganddrought
mediatethefollowing:(1)adeclineinthetaxonomicandfunctional
diversityoftraitsand(2)ashiftinthemeanfunctionaltraitvalues
asdepicted inFigure1. Theseresponses will expectedly bestron-
gest in nea r-stream a reas where the hydrol ogical alterati ons are
most pronounced and willintensify over time.Additionally,itwas
testedif(3)thetaxonomicdiversityandfunctionaldiversityofthe
seed poo l were higher in fl ooded areas th an in drought area s as
theregional speciespoolmay contributeto diversitythroughspe-
cies dispersal by water (i.e., hydrochory; Nilsson, Brown,Jansson,
&Merritt,2010).
2 | MATERIALS AND METHODS
2.1 | Experimental setup
Four riparian areas situated along streams in Denmark
(Sandemandsbækken 56.158507N, 9.496120 E; Voel Bæk
56.195846N, 9.703932 E), Germany (Boye 51°58′61.1″N,
6°91′10.01″E), and the Netherlands (Groote Molenbeek
51°39′17.32″N, 6°03′59.47″E)were selected fortheexperiment
(Table1).Thefourstreamsvariedinmeandischargefrom0.03to
1.73m3/s.This was, however,notconsideredproblematic asour
FIGURE1 Hypothesizedchangesincommunitytrait
compositionmovingfromdroughttofloodedconditions.Arrows
indicatewhetheratraitisexpectedtoincreaseordecreasewith
increasedflooding,withanexpectationoftheoppositeresponseto
drought
Regenerative
• Seed mass (SM) ↓
• Seed buoyancy (BYC) ↑
• Seed volume (SV) ↑
Vegetative
• Specific leaf area (SLA) ↑
• Leaf size (LS) ↑
• Leaf mass (LM) ↓
• Canopy height (CH) ↑
• Root porosity (RP) ↑
• Rooting depth (RD) ↓
DroughtFlooded
Site Sandemandsbæk Boye Voel Bæk Groote Molenbeek
Catchmentarea
(km2)
0.07 3.40 7. 57 183.56
Grassland(%) 0.16 0.31 0.02 0.43
Forest(%) 0.43 0.11 0.03 0.00
Urban(%) 0.05 0.15 0.04 0.07
Agriculture(%) 0.25 0.42 0.90 0.45
Wetlands(%) 0 .11 0.00 0.00 0.00
Water(%) 0.00 0.01 0.00 0.00
Meandischarge
(m3/s)
0.03 0.08 0.06 1.73
TABLE1 Studysitecharacteristics
4
|
BAATTRUP- PEDERSEN E T Al.
samplingeffortwasfocused on coveringthe naturalfeaturesof
thest re am-rip ar iangr ad ientatthes tudysitesirr es pect iveofsize.
Thatis,thesamplingcoveredagradient from the watertableof
the stre am under sum mer base flow co nditions to t he high end
of the floodplain where only extreme events lead to flooding
(Figure2).
Thelengthoftheexperimentalareaswas150m,whereaswidth
varieddependingontheextentofthestream-ripariangradient.The
areas were dividedintothree sections: acontrol section, a winter/
spring fl ooded sec tion, and a summ er drought sec tion. These hy-
drologicaltreatmentswereselectedtomimichydrologicalchanges
inEurope as predictedbyIPCC (2007).The riparian areas hadnot
been exposed to floodings prior to theexperiment and comprised
seminaturalgrasslandcommunitieswithonlyherbaceousspecies.
The control sections were situated upstream of the manipu-
latedsectionswithbufferareasin-between(Figure2).Floodingwas
created byconstructingdamsin the streamstoobstructthewater
flowinthemainchannels.InDenmark,alateraldammadeofsand-
bagswasestablishedacrossthestreamchannel(Figure2a),whilein
Germanyand theNetherlands,longitudinaldamswerebuilt within
the channel, which together withalateral damacross thechannel
obstructedthewaterflowinthemainchannel(Figure2b).Thecon-
structeddamswereusedtocreatea6-weekfloodingoftheadjacent
riparianareas (from March to mid-April) in 2011,2012, and2013,
where the strongestresponseswereexpectedto occurin the final
year ofsamplinggiven that theareas have been subject to manip-
ulation fo r several years . However, in 2013, floodi ng was delayed
in Denmar k due to ice cover and las ted from the end of A pril to
mid-June. In D enmark, summ er droughts we re created by digg ing
aditch, which together with a lateral dam inthe mainchanneldi-
verted p art of the wate r flow from the mai n channel, re sulting in
alowered water tablewithin the experimentalareas (Figure2a).In
Germany and the Netherlands,alongitudinaldamwasconstructed
acrossthestreamchannel,whichtogetherwithalateraldamacross
thechannelobstructedthe waterflowadjacenttothe experimen-
tal area, thereby loweringthewater table(Figure2b). The drought
experiment was conducted in 2011, 2012, and2013from the end
of June to Sep tember (appr oximately 10week s)at a ll sites except
Boyewherestronggroundwaterseepagepreventedreductioninthe
watertableinthedroughtsection.
Within e ach section, t hree sample tr ansects wer e establishe d
perpendicular tothe stream from the channel andupwards in the
riparianareas(Figure2). Thelength of the sampletransec ts varied
amongthestudysitesinordertorepresentagradientfromthelow-
est watertable ofthe stream under summer base flow conditions
tothehighestpoint ofthestreamvalleypotentiallyfloodedbysur-
face waterduring extremewinter floods (Figure2c). Todetermine
thehydrology ofthecontrol,drought,and floodedsections,atotal
ofnine piezometerswereinstalledwithineachsection(threealong
eachsampletransect).Thefirstpiezometerwasplacedclosetothe
stream , just above the n ormal summer w ater table in th e stream,
thatis normallynotfloodedduringsummerbutoccasionallyduring
winter flo ods (position 1; Figu re2c). The second piezometer was
placedjustabovethenormalwinterwatertablethatisnormallynot
floodedineithersummerorwinter(position2;Figure2c).Thethird
piezomete r was placed at th e highest poin t of the floodpl ain that
was rarely flooded and, ifso, only during extreme winter flooding
events(onceevery100years;position3;Figure2c).
2.2 | Characterization of hydrology and vegetation
The water t able depths we re measured at l east four tim es during
theexperimental periodsineachexperimental year (atthestartof
theexperiment,after2weeks,after4weeks,andattheendofthe
FIGURE2 Aschematicpresentationoftheexperimentalsetup
appliedinourstudy.Thecontrolsectionissituatedupstreamofthe
droughtandfloodedsectionswithbuf fersin-between.Flooding
wascreatedbyconstructingdams(markedasbarsonthefigure)
toobstructthewaterflowinthemainchannels.(a)InDenmark,
alateraldamofsandbagswasconstructedacrossthestream
channel.(b)InGermanyandtheNetherlands,longitudinaldams
werebuiltwithinthechannel,whichtogetherwithalateraldam
acrossthechannelobstructedthewaterflowinthemainchannel.
(c)Thepositionofthesampletransectswithintheexperimental
sections.Thefirstpiezometerwasplacedjustabovethesummer
watertable(position1),thesecondpiezometerjustabovethe
normalwinterwatertable(position2)andthethirdatthehighend
ofthefloodplain(position3).Thecirclesindicatethepositionofthe
piezometersalongeachtransect
SWT
1
NWT
2
HFP
3
(a) (b)
(c)
Buffer
ControlDrought
Buffer
Buffer
Flooded
ControlDroughtFlooded
Buffer
|
5
BAATTR UP- PEDERSEN E T Al.
experiment).MeanvaluesofwatertabledepthsaregiveninTable2.
Positive va lues indicate t hat flooding occ urred; the more p ositive
thevalues,thehigherthefloodingdepths.Similarly,negativevalues
indicatethat the water tableis situatedbelowthesurface,andthe
morenegativethe values,thedeeperthe watertable.Close tothe
streams(position1),thefloodingtreatmentprolongedtheduration
ofwinterfloodingandincreasedthedepthofflooding,whereasthe
drought treatment generally lowered the groundwater table dur-
ing the tre atment perio d (Table2). Further away f rom the strea m
at position 2, theflooding treatment resulted in occasional winter
floodingsduring the treatmentperiod, whereas thedroughttreat-
mentlowered the groundwater table( Table2).Far thestaway from
the stre am (position 3), t he flooding tr eatment result ed in overall
highergroundwatertablesduringthetreatmentperiod,whereasthe
droughttreatmentloweredthegroundwatertable(Table2).
Vegetation surveyswereconductedduring the growingseason
(June–September).Percentagecoveragewasestimatedforallvascu-
larspeciesinatotalof27plots(50×50cm2)persiteforeachtreat-
ment. These were positioned with three plots next toeach of the
threepiezometersineachofthethreetransects.Speciescomposi-
tionwasrecordedaccording totheBraun-Blanquet method(1928),
adjuste d by Barkman, D oing, and Seg al (1964). In the t wo Danish
sites, anadditional27bareplotswereestablished with threeplots
nexttoeachofthethreepiezometersineachofthethreetransects
inordertofollowtheestablishmentofthevegetationunderthenew
hydrologic al settings d uring the exper imental perio d. These were
Site Treatment Position
Groundwater,
mean (cm) Groundwater, SE
Sandemandsbækken Control 1−10 . 35 1.68
2−22 .35 1.73
3−16 . 59 1.32
Drought 1−18.86 1.44
2−26 .75 1.89
3−20 .9 3 2.45
Flooded 11.34 1.9 0
2−0.77 2.27
3−26 .4 3 1.07
Voel Control 1−10.07 0.94
2−16 .1 3 1.01
3−29. 36 1.56
Drought 1−35.23 1.73
2−4 9.3 5 2.20
3−56.10 2.39
Flooded 11.10 1.79
2−0.50 1.77
3−24.2 8 1.63
Boye Control 1−8.79 1.81
2−9.9 6 2.13
3−22 .74 3.36
Flooded 113.70 2 .10
2−0.18 3.68
3−30.12 2.12
GrooteMolenbeek Control 1−5. 27 3.09
2−15 . 87 2.53
3−21.5 2 3.48
Drought 1−8.72 1.78
2−33.0 0 2.14
3−37.75 3.14
Flooded 113.55 4.62
21.29 2.89
3−4.50 1.39
TABLE2 MeansandSEof
groundwatertabledepthsmeasuredin
piezometersatleastfourtimesduring
eachexperimentalrun(atthestartofthe
experiment,after2weeks,after4weeks,
andattheendoftheexperiment).Positive
valuesindicatethatthewatertablewas
situatedabovethegroundsurface,and
negativevaluesindicatethatthewater
tablewassituatedbelowtheground
surface.Thepiezometerswereplaced
alongahydrologicalgradient.Thefirst
samplingpointwasatthelowestwater
tableofthestreamduringsummerbase
flowconditions(SWT).Thesecond
samplingpointwasjustabovethenormal
winterwatertablethatisnormallynot
floodedineithersummerorwinter
(position2).Thethirdsamplingpointwas
atthehighestpointupthestreamvalley
thatcouldbefloodedbysurfacewater
duringextremewinterfloods(position3)
6
|
BAATTRUP- PEDERSEN E T Al.
createdbyremovingtheexistingvegetationandthetopsoilfollowed
bydepositionof 15cm mixedsandand peat.Toavoidingrowthof
nearbyplants, the plots were delineated using15-cm-wide plastic
bandsthatwereverticallyinsertedintothesoil.
Vegetationdatawereconver tedtoOrd%scale(coverageranges
from 0.5 t o 140) according to Van de r Maarel (20 07) for a cover-
basedinterpretationoftheBraun-Blanquetscale(Braun-Blanquet,
1928). Seed traps consisting of 25×22.5cm artificial mats with
plasticbristles(Astroturf®)wereplacedandsecurednearthesquare
plots used for vegetation sur veys. Seeds were collected in 2011
in both control, flooded, and drought areas duringthe 6weeks of
experimental flooding and10weeks of experimental drought. The
matswereremovedfromthefieldimmediatelyaftertheexperimen-
tal per iod and taken to t he laborator y where the y were stored in
plastic bags i n the dark at 4°C bef ore processing . The processin g
involvedextractionofdepositedmaterialbyflushingtheseedtraps
withwater,followed bywet sieving thedepositstoremovefinesilt
andclay.Thematerialwasthendriedat70°Cfor48hrafterwhich
intactseeds werevisually identified from thedried material, man-
uallyremoved,anddetermined tospecieslevelwiththeuseof the
“Digital seed atlas of the Netherlands” (C appers, Bekker, & Jans,
2006).
2.3 | Diversity indices and community- weighted
means of plant traits
All diversity and trait indices were calculated for each vegetation
type based on Ord%values (van der Maarel 2007). Wecalculated
taxonrichnessandShannondiversityasindicesoftaxonomicdiver-
sity. Tra it swerealloca te dt ot he en co un te redspecie sb as ed on in fo r-
mationavailable intheLEDAdatabase(Kleyer&Bekker,2008)and
literaturecitedinDoumaetal.(2012).Weselectedtraitsdescribing
both see d (SM, BYC, SV; Table3) and adu lt (SLA , LS, LM, C H, RP,
RD;Table3)plantcharacteristicsexpectedtorespondtoanaltered
hydrologicalregimeasdescribed intheintroduction(Figure1). The
number of specieswithtrait information and thetotalabundances
ofthese species are given in Table3. We calculated functional di-
vergence (FDvar) and community-weighted means (CWMs) when
the abun dance of species wit h trait informatio n was above 65%,
thereby precluding specific leaf area, root porosity, and rooting
depth(Table3).Theabundancelimitrepresentedabalancebet ween
ontheonehandtohaveasmanytraitsaspossibleintegratedinthe
analysesto obtaininsight intothe functional responseoftheplant
communit ytoclimatechange-relatedalterationsinthehydrologyof
the areas , and on the oth er hand to keep the e stimation b ias low
(Borgyetal.,2017).FDvarandCWMswerecalculatedforeachtrait
accordingtoLavoreletal.(2007).
Aresponse ratio (Δr) (Osenberg, Sarnelle,& Cooper,1997) for
eachdiversityandtraitmetricwasalsocalculatedusingmeanvalues
ofthree sample plot s for eachofthe threesamplingtransects for
eachpositionas:
where Ncisthemeanmetricvalue atthecontrolsiteandNtis the
metricvalueforthetreatment(floodedordrought).Responseratios
allowedustoassessthegeneraleffectsofthetwotreatmentsonri-
parianplantdiversityandtraitcompositionacrossthefourstreams.
2.4 | Data analyses
Allanalysesdescribedinthisparagraphwereconductedusingthesta-
tisticalsoftwareR(RCoreTeam2014),packagevegan(Oksanenetal.,
2014). Canonical correspondence analysis (CCA) (function cca) fol-
lowedbypermutationalANOVAs(function anova.ccawithmaximum
permutations set to 9999) was performed to assess differences in
plantcommunity compositionbetweentreatments (control,drought,
flooding),typeofvegetation(seed,existingvegetation,bareplot),and
year(2011,2012,2013).Toestimatetheuniqueeffectofasinglepre-
dictor (i.e., treatment,type of vegetation, and year),the variation in
plantcommunity compositionexplainedbythe otherpredictors was
alwayspartialledout(i.e.,includedascovariables)intheANOVAs.We
also assessed which traits weresignificantly associated with differ-
encesinplantcommunitycompositionbetweentreatmentsbyfitting
traitvectors(describingtherelativeabundanceoftraitsineachplot;
i.e.,CWMs)ontotheCCAordinationusingthefunctionenvfit. The en-
vfitfunctionfindsthedirectionintheordinationspacetowardwhich
eachtraitvectorchangesmostrapidlyandtowhichitismaximallycor-
relatedwiththeordinationconfiguration.Thesignificanceofthetrait
vectorswasdeterminedbyapermutationtest(n=999).
Δ
r=ln
(
Nt
Nc
)
TABLE3 Explanationsofthetraitsusedtocharacterizethe
riparianplantcommunities.TraitswerederivedfromtheLEDA
database(Kleyer&Bekker,2008)andfromliteraturecitedin
Doumaetal.(2012).Thepercentageofspecieswithtrait
informationwascalculatedasthenumberofspecieswithtrait
informationandastheabundanceofspecieswithtraitinformation
(inbrackets).Threetraitswereexcludedfromtheanalyses(SLA,
RD,RP)astheabundanceofspecieswithtraitinformationwas
below65%
Trait name Unit Category
% species with
trait information
Seedbuoyancy
(BYC)
%Seed 64(65)
Seedmass(SM) Mg Seed 75(78)
Seedvolume
(SV)
mm3Seed 68(73)
Specificleafarea
(SLA)
mm2/mg Adult 52(55)
Leafsize(LS) mm2Adult 64(70)
Leafmass(LM) Mg Adult 62(68)
Canopyheight
(CH)
MAdult 74(77)
Rootporosity
(RP)
%Adult 31(53)
Rootingdepth
(RD)
MAdult 37(65)
|
7
BAATTR UP- PEDERSEN E T Al.
Toassessthegeneraleffectsofthetreatmentsacrossthestudy
streams,we combinedtheyearlyestimatesintoasingle effectsize
measurementand tested whether the response ratios (Δr) of tax-
onomic diversity, trait diversity, and CWMs differed significantly
from zero (i .e., higher or lowe r than zero) using t wo-side d ttests.
The yearly response ratioestimateswerecombinedby a weighted
averageusingthevariancefor yearastheweight.Ttestswereper-
formed separately foreachvegetationtype (seed,existingvegeta-
tion, bareplot). Asignificant result was interpreted as a consistent
anddetectablechangeinthemetricvalueinthecontrolsiteversus
thetreated(floodedordry)siteacrosstheinvestigatedstreams.
3 | RESULTS
There were large variationsinspecies composition amongthefour
studysitesregardingbothtypeconsidered(i.e.,seedpool,bareplot,
and existing vegetation), treat ment applied (i.e., control, drought,
and flooding), and time of sampling (i.e., 2011, 2012 and 2013;
Figures3 a nd 4; Table4). The effec ts of the ap plied treat ment on
the compositionalpatterns in the experimentalareas were signifi-
cant for both the seed pooland the existing vegetation(Figures3
and4;Table4).Severalofthetraitsusedtodescribethefunctional
characteristics of the vegetation were associated with the main
FIGURE3 Ordinationplotsofthecanonicalcorrespondenceanalyses(CCAs)ofplantspeciescompositionwithineachriparianarea
(Boye,GrooteMolenbeek,Sandemandsbækken,andVoelBæk).IntheCCAs,speciescompositionwasconstrainedbythetypeofvegetation
(seed,existing,andbareplot),whereasthevariationinspeciescompositionexplainedbytreatment(flood,drought,control)andyear(2011,
2012,2013)waspartialledout.TraitssignificantlyassociatedwiththeCCAaxes(p<.05)areplottedontotheordination
–1 012 43
–3
–2
–1
0
1
2
3
CCA1
CA1
Control
Flooded
Boye
–2–3 –10 123
–3
–2
–1
0
1
2
3
CA1
–6 –4 –2 0246
–6
–8
–4
–2
0
2
4
CCA1
Sandemandsbækken
–3–4 –2 –1 012
–3
–4
–2
–1
0
1
2
CCA1
CCA2
Voel Bæk
Groote Molenbeek
CCA1
Control
Drought
Flooded
Existing
Seed
SM
CH
LM
LS
Existing
Seed
SM
CH
LM
LS
CCA2
Control
Drought
Flooded
Exis ting
Seed
Bareplot
SM
CH LS Exis ting
Seed
Bareplot
BYC
SM
LS
Control
Drought
Flooded
8
|
BAATTRUP- PEDERSEN E T Al.
gradientsintaxonomiccomposition(Tables5and6),suggestingthat
theycapturedimportantunderlyingmechanismsresponsibleforthe
observedcompositionalchanges.
3.1 | Existing vegetation
Apply ing response rat ios, we detecte d consistent chan ges among
studysitesforboththetaxonomicandfunctionalcompositionofthe
plantcommunities. Inaccordancewiththefirst hypothesis,weob-
servedthatbothspeciesrichnessandShannondiversitywerenega-
tivelyaffectedbydroughtandfloodingandthattheresponsevaried
with dist ance from the s treams (Fig ure5). At position 1 , the rich-
nessanddiversityoftheexistingvegetationdeclinedinresponseto
droughtthefirstyearafterinitiatingthetreatment(i.e.,theresponse
ratiowassignificantlylowerthanzero),andrichness wasstilllower
after3yearsoftreatment(Figure5).Fur therawayfromthestreams
atposition 2,we observeda decline in species richnessand diver-
sity,buttheresponse was only significant after3yearsof flooding
(Figure5).
Inaccordancewiththe secondhypothesis,wealsoidentified
con siste ntcha ngesi nthef un ctio naldi versityoftheexistingvege -
tationinparticularinresponsetoflooding(Figures6a,7a,and8a).
FIGURE4 Ordinationplotsofthecanonicalcorrespondenceanalyses(CCAs)ofplantspeciescompositionwithineachstream(Boye,
GrooteMolenbeek,Sandemandsbækken,andVoelBæk).IntheCCAs,speciescompositionwasconstrainedbytreatment(flood,drought,
control),whereasthevariationinspeciescompositionexplainedbytypeofvegetation(seed,existing,andbareplot)andyear(2011,2012,
2013)waspartialledout.TraitvectorssignificantlyassociatedwiththeCCAaxes(p<.05)areplottedontotheordination
–3 –2 –1 0123
–3
–2
–1
0
1
3
2
CA1
Boye
–2–3 –1 01
23
–2
–1
0
1
2
4
3
CCA1
CCA2
Groote Molenbeek
–4 –2 024–4–20
24
–1
–3
–2
0
1
2
3
5
4
–1
–3
–2
0
1
2
3
5
4
CCA1
CCA2
Sandemandsbækken
CCA1
CCA2
Voel Bæk
CCA1
Existing
Seed
Control
Flooded
BYC
SM
LM
LS
Control
Drought
Flooded
BYC
LS
Existing
Seed
Control
Drought
Flooded
Existing
Bareplot
Seed
SM
SV Control
Drought
Flooded CH
LM
Existing
Bareplot
Seed
|
9
BAATTR UP- PEDERSEN E T Al.
Close tothestreams,atpositions 1 and 2, weobservedthatthe
functionaldiversityofalltrait sdeclinedinresponseto3yearsof
flooding(BYCSM,SV,CH,LM,andL S;Figures6aand7a),whereas
thefunctional diversity of CHdeclinedinresponse to 3years to
drought bu t only at position 1 (cl osest to the str eam). Farthest
away from th e streams at posit ion 3, we observe d a decline in
the functional diversityoftwo trait s (LM and LS)inresponseto
drought(Figure8a).
In accordance with the second hypothesis, we also obser ved
consisten t changes in the mean f unctional tr ait (CWM) values of
Constraint Covariables Study site X2F (df)Pr (>F)
Treatment Type;Year Boye 0.352 2.571(1.21) 0.005
GrooteMolenbeek 0.537 2.847(2.3 4) 0.005
Voel 0.377 2.976(2.58) 0.005
Sandemand 0.432 2.682(2.58) 0.005
Typ e Treatment;Year Boye 0.909 6.6 47(1 .21) 0.005
GrooteMolenbeek 0.642 6.798(1.34) 0.005
Voel 0.680 5.409(2.58) 0.005
Sandemand 0.842 5.221(2.58) 0.005
Yea r Treatment;Type Boye 0.352 1.224(2.20) 0.079
GrooteMolenbeek 0.409 2.0 84(2.34) 0.005
Voel 0 .224 1.696(2.58) 0.005
Sandemand 0.269 1.614(2.58) 0.005
TABLE4 Summarystatisticsofthe
ANOVAsofthecanonicalcorrespondence
analyseswherespeciescompositionwas
constrainedbytreatment,t ype,oryear.
Thevariationoftheotherparameterswas
alwayspar tialledout(i.e.,includedas
covariables)intheANOVAstoenable
estimationoftheuniqueeffectofasingle
parameter
TABLE5 Summarystatisticsoftheenvfitanalyseswheretraitvectors(CWMs)werefittedtotheordinationaxesofthecanonical
correspondenceanalyses(CCAs).Summarystatisticsofthecorrelationbetweentraitvectorsandthefirstt woordinationaxesareshown.In
theCCAs,plantspeciescompositionwasconstrainedbythet ypeofvegetation,whiletreatmentandyearwereincludedascovariables(i.e.,
thevariationinplantcompositionexplainedbytreatmentandyearwaspartialledout)
Tra it
Boye Groote Molenbeek Sandemandsbæk Voel Bæk
CCA1 CA1 r2CCA1 CA1 r2CCA1 CCA2 r2CCA1 CCA2 r2
BYC −0.07 1.00 .05 0.80 0.60 .06 −0.53 0.85 .07 −0.44 0.90 .21**
SM 0.78 0.63 .30**** −0.52 −0.85 .08 0.95 0.33 .00 0.84 0.54 .18**
SV 0.42 0.91 .10 − 0.62 −0.79 .02 0.50 −0.87 .02 0.54 0.84 .06
LS 0.99 0.17 .32** −1.0 0 −0.03 .21*−0.99 0.12 .12*−1.0 0 −0.07 .15*
LM 0.85 −0 .52 .63** −0 .98 −0 .19 .40*** 0.58 0.81 .01 −0.89 −0.46 .07
CH 0.99 −0.13 .88*** 0.60 −0.80 . 26** −1. 0 0 0.08 .45*** −0.68 0.73 .08****
***p<.001,**p<.01,*p < .05.
TABLE6 Summarystatisticsoftheenvfitanalyseswheretraitvectors(CWMs)werefittedtotheordinationaxesofthecanonical
correspondenceanalyses(CCAs).Summarystatisticsofthecorrelationbetweentraitvectorsandthefirstt woordinationaxesareshown.In
theCCAs,plantspeciescompositionwasconstrainedbytreatment,whilethetypeofvegetationandyearwereincludedascovariables(i.e.,
thevariationinplantcompositionexplainedbytreatmentandyearwaspartialledout)
Tra it
Boye Groote Molenbeek Sandemandbæk Voel Bæk
CCA1 CA1 r2CC A1 CA1 r2CCA1 CCA2 r2CCA1 CC A2 r2
BYC −0.89 0.46 .25 −0 .41 0.91 .32 0.66 0.75 .07 −0.02 1.00 .07
SM −0.36 0.93 .14 0.67 −0.75 .02 1.00 −0.09 .14*0.94 −0.35 .01
SV 0.53 0.85 .11 −0.23 −0 .97 .02 0.91 −0.40 .17** −0.70 −0.72 .07
LS 0.98 0.18 .31*−1. 0 0 −0.04 . 24*0.99 −0.11 .03 −0.15 0.99 .05
LM 0.77 −0.63 .43** −0.78 −0.63 .15**** − 0.81 − 0.59 .04 0.64 0.77 .11*
CH 0.54 −0.84 .02 −0.83 0 .55 .13**** 0.83 0.55 .05 0 .47 0.88 .16**
***p<.001,**p<.01,*p < .05.
10
|
BAATTRUP- PEDERSEN E T Al.
the existing vegetation in response to the applied treatmentsand,
asdemonstratedbythediversitypatterns,theresponsevariedwith
distancefromthestreams(Figures6b,7b,and8b)andgenerallyfol-
lowed the pr edicted pa tterns (se e Figure1). Close t o the stream s,
atposition1,BYC-CWMincreasedinresponsetofloodingandSM-
CWMincreasedinresponse todrought (Figure6b)but, incontrast
to our expectations, SV-CWM declined in response to flooding.
Furthe r away from the st ream, at posit ion 2, BYC-CWM an d CH-
CWM increased in response to flooding and LS-CWM declined
(Figure7b), but incontrast toourexpectations,LM-CMW declined
inresponsetodrought (Figure7b). Farthest away from thestream
atposition 3,weobservedanincreaseinSV-CWM inresponse to
flooding,alsoconfirmingourexpectations(Figure8b),butSM-CWM
increasedwhichwasincontrasttoourexpectations(Figure8b).We
alsoobservedseveralsignificantchangesinthetraitcompositionof
thecommunit yinresponsetodroughtatposition 3(BYC,SM,CH,
LS,SV,LM)andforthemajorityofthetraits,thesechangeswereas
predicted(BYC,SM,CH,LS;Figure8b).
3.2 | Seed pool
As oppo sed to our third hyp othesis, we did not f ind a signific ant
increaseinthetaxonomicrichnessordiversit yofthe seed poolin
respons e to flooding ( Figure5; ANOVA; p>.05), b ut we obser ved
an increase in functional diversity but only for SM at position 2
(Figure7a).Instead,weobservedseveralchangesin the traitvalue
oftheseedpool inresponseto flooding (CH, LM, LS at position1;
SM,LM,LSatposition2;BYC,SM,SV,CH,LM,LSatposition3)and
drought(CH,LM,LSatposition1;CH,LSatposition2)andmostof
thesechangesfollowedthepredic tedpat tern(Figure1)particularly
closetothestream.
4 | DISCUSSION
4.1 | Taxonomic and functional diversity response
Wefoundsignificanteffectsoffloodinganddroughtonthespecies
composition of both the vegetation and the seedpool in allstudy
areas. B etween-study si te variabilit y was also pro minent, an d this
islikelyduetolocaldif ferencesinsoilcharacteristicsand/orhydro-
logical conditionsamong the study sites thatinfluencethe effects
ofhydrologicalalterationsontheriparianvegetation(Garssenetal.,
2015).Despitetheobservedbetween-studysitevariability,consist-
entpatternswerealsodetectedinresponsetohydrologicalchanges.
Inparticular,weobservedadeclineinboththetaxonomicandfunc-
tional diversityoftheplantcommunities.Thedecline intaxonomic
diversit yinresponsetodroughtwasonlyevidentnearthestreams,
probabl y reflecti ng that the expe rimental ar eas were alread y well
FIGURE5 Averageresponseratios(±1SE)oftaxonomicdiversity(richnessandShannondiversity)inplotspositionedclosetothestream
channeljustabovethenormalsummerwatertable(position1;a)andinplotssituatedjustabovethenormalwinterwatertable(position2;
b).Nosignificantchangesinrichnessordiversityoccurredfurtherupthefloodplain,position3,followingtheapplieddroughtandflooding
treatment.Opensymbols(existing)comprisedataforthevegetationsurveys,whereasclosedsymbols(seed)comprisedatafortheseed
trapsur veys.Thecoloroftheasteriskindicatesthetypeofvegetationdifferingsignificantlyfromzero(i.e.,blackasterisk=seed,white
asterisk=existing)
RichnessShannon
–1.0
–0.5
0
0.5
1.0
–2.0
–1.0
0
1.0
2011 2012 2013 2011 2012 2013
Drought Flooded
Existing Seed
(b)(a)
2011 2012 2013 2011 2012 2013
Drought Flooded
RichnessShannon
–1.0
–0.5
0
0.5
–2
–3
–1
1
0
|
11
BAATTR UP- PEDERSEN E T Al.
drainedandconsequentlylessaffectedbytheexperiment( Table2),
whereasthenegativeimpactsoffloodingonspeciesdiversitywere
more pronounced (although only significantafter 3years offlood-
ing).Thisfindingmayindicatethatfewerspecieswereabletotoler-
atefloodingwithin the areacompared withthenumber of species
able to tolerate(relativelymild) droughtand/orthatdispersalcon-
straintswerehigherforspeciesadaptedtofloodedconditions.Our
findingsareinlinewiththoseofStröm,Jansson,Nilsson,Johansson,
and Xiong(2011)where soil monolithswere transplantedto areas
subjectedtodifferentfloodingintensitieswithintheriparianzoneof
aborealriver.Speciesdiversityincreasedrapidlyinmonolithstrans-
plantedtohigherelevations(i.e.,lessflooding)overthecourseofthe
6-year fieldstudy,whilespeciesdiversityinmonolithstransplanted
tolowerelevations(i.e.,moreflooding)declinedrapidly(Strömetal.,
2011).
Functional diversity also respondedto the altered hydrological
settings,inparticularinproximitytothestreams.Weobservedasig-
nificantdeclineinthefunctionaldiversityofalltraits,indicatingthat
the range ofsuccessful strategies displayed underthe new hydro-
logicalsettingswasrestricted.Ourfindinglendssupporttoprevious
studiessuggestingthatstrong abioticfilters constraintherange of
speciesmeantraitvaluesthatcanexistwithinthecommunity,lead-
ing to a convergent traitdistribution (Bernard-Verdier etal., 2012;
Jung, Violle, Mondy,Hoffmann, & Muller,2010;Weiher,Clarke, &
Keddy,1998).In linewithourobservationsfor taxonomicdiversity,
also functional diversity responded morestronglyto flooding than
drought,indicating thatfloodingposesamoreseverestress onthe
riparian community intemperate regions (Fraaije,B raak, Verduyn,
Verhoeven,&Soons,2015;Fraaije,Braak,Verduyn,Breeman,etal.,
2015).Theloss offunctionaldiversity(1–2years)may influencere-
sourceuseefficiencywithinthesystems,withcascadingeffectson
ecosystem functioning (Díaz & Cabido, 2001). Further studies are,
however, needed to explore this topic, with special emphasis on
how climate change-mediatedalterations in hydrologicalextremes
incombinationwithahigherdegreeofunpredictabilityintheoccur-
renceoftheseaffectecosystemfunctioning.
4.2 | Community functional trait response
Thelossof functional diversitywasalsoreflected inthemeantrait
responseoftheriparianplantcommunit y.Weobservedaconsistent
FIGURE6 Averageresponseratios(±1SE)offunctionaltrait
diversit y(FDis)(a)andtraitcomposition(CWMs)(b)inplots
positionedclosetothestreamchanneljustabovethenormal
summerwatertable(position1).Whenaresponseratiois
significantlydif ferentfromzero,thisisindicatedwithanasterisk
abovetheerrorbar(p<.05).Opensymbols(existing)comprisedata
forthevegetationsur veys,whereasclosedsymbols(seed)comprise
datafortheseedtrapsurveys.Thecoloroftheasteriskindicates
thetypeofvegetationdifferingsignificantlyfromzero(i.e.,black
asterisk=seed,whiteasterisk=existing).Notethatthescalefor
FDisforCHisdif ferentincomparisonwiththeothertraits
Existing Seed
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–4
–2
–6
6
4
2
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
Drought Flooded
BYCSMSVCHLM
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
LS
DroughtFlooded
(a) Functional divergence (b) CWM
0
–4
–2
–6
6
4
2
12
|
BAATTRUP- PEDERSEN E T Al.
increas e in the mean tra it value of seed b uoyancy in res ponse to
flooding,indicatingthatthefractionofspeciesadaptedto flooded
conditionsincreasedin thearea.This findingisin accordance with
Ozinga,Bekker,Schaminee,andVanGroenendael(2004)who,based
onaclassificationofdispersaltraitsofca.900speciesfromdifferent
typesofcommunities,foundahighlysignificantcorrelationbetween
thepositionofspeciesalongawetnessgradientandthe frequency
ofmorphological adaptationsto hydrochory.This pat tern has later
beenconfirmedalsoforriparianandaquaticplantcommunities(van
denBroek,vanDiggelen,&Bobbink,2005).Asopposedtothefind-
ings ofDouma etal. (2012), however,we did notobserveadeclin-
ingseedmasswith enhancedbuoyancyandseeddensitytherefore
seemstobearelativelypoorpredictorofseedbuoyancy.
For the vegetative CWMs, we obser ved fewer consistent
changesincomparisonwiththosepreviouslyreportedtorespondto
analteredhydrology(Bernard-Verdieretal.,2012;Jungetal.,2010;
Mommer,De Kroon,Pierik,Bögemann, &Visser,2005;Violleetal.,
2011;Voesenek,Colmer,Pierik,Millenaar,&Peeters,2006).There
may be several, nonmutually exclusive, explanations to the less
consistentresponse of trait CWMs to thecontrasting hydrological
settingsin our study.First, differentadaptive strategiesfordiffer-
entspeciesmayco-occurinacommunity,whichmay partlyexplain
therelatively weak responseobservedwhencomparing the mean
traitvalueofsingletraits(Bernard-Verdieretal.,2012;Doumaetal.,
2012). For exam ple, some sp ecies may have smal l and thin leaves
that facilitate oxygen uptake during submergence (Banach etal.,
2009;Nielsen&Sand-Jensen,1989),enablingthemtosurviveunder
floodedconditions,whereas otherspeciesmay avoidflooded con-
ditions by e longating the ir shoots, th ereby accessing at mospheric
oxygen ( Voesene k, Rijnders, P eeters, Van de Steeg, & D e Kroon,
2004)as alsoobservedinourstudy. Second,intraspecific variabil-
itywasnotconsideredforanyofthetrait sinthisstudy,whichmay
have weakened community responses (Albert, Grassein, Schurr,
Vieille dent, & Violle, 2011; Jun g etal., 2010). For example , many
speciescanelongateshootsandpetiolesthatenablethemtosurvive
shallow,prolongedflooding(e.g.,Chenetal.,2009),butsuchabilities
willnotbecapturedwhenapplyingmeantraitvalues.Third,weonly
followed thecommunitiesfor3yearsafterthechangeinhydrolog-
icalsettings.Alteredhydrologicalconditionswill likelymediate fast
exclusionofspeciesintolerantofthesechanges,whereastheestab-
lishment ofnewspecies relieson theirdispersal and establishment
FIGURE7 Averageresponseratios(±1SE)offunctionaltrait
diversit y(FDis)(a)andtraitcomposition(CWMs)(b)inplots
positionedjustabovethenormalwinterwatertable(position2).
Whenaresponseratioissignificantlydifferentfromzero,this
isindicatedwithanasteriskabovetheerrorbar(p<.05).Open
symbols(existing)comprisedataforthevegetationsurveys,
whereasclosedsymbols(seed)comprisedatafortheseedtrap
surveys.Thecoloroftheasteriskindicatesthetypeofvegetation
differingsignificantlyfromzero(i.e.,blackasterisk=seed,white
asterisk=existing.NotethatthescaleforFDisforSMisdifferent
incomparisonwiththeothertraits
Existing Seed
0
–2
–1
–3
3
2
1
0
–2
–1
–4
–3
4
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
Drought Flooded
BYCSMSVCHLM
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
LS
DroughtFlooded
(a) Functional divergence (b) CWM
0
–2
–1
–4
–3
4
3
2
1
|
13
BAATTR UP- PEDERSEN E T Al.
within th e areas. The refore, a delay i n the respon se of mean trai t
values of thecommunit y to changedhabitat conditions mayoccur
(Oddershede,Svenning,&Damgaard,2015;Sandeletal.,2010),re-
flectingprogressivefillingofavailablenicheswithinthecommunity,
eventuallyleadingtostronger traitconvergence (Helsen,Hermy,&
Honnay,2012;Roscher,Schumacher,Gerighausen,&Schmid,2014).
Thisdelaymaybestrongerinexistingvegetationthan inbareplots
where col onization and envi ronmental filt ering may occur rapi dly
(Fraaije, Braak, Verduyn, Verhoeven, etal., 2015; Fraaije, Braak,
Verduyn, Br eeman, etal., 2015) as a lso seen in the bare pl ots in
ourstudy, which dif fered significantlyinspeciescompositionfrom
the exis ting vegetatio n. Finally, we did not h ave traits for a ll spe-
cies foun d in the areas, a nd the result s regardi ng the respon se of
communit y-weightedtrait meansshould therefore be treated with
caution.
4.3 | Seeds
Weexpected to find functionallymorediverseseedpools inthe
floodedareasthaninthedroughtareas,reflectingthathydrochory
canintroduceseedsfromanupstreamspeciespoolinadditionto
seeds that may enter from the localspecies poolby wind and/or
animal dispersal.Furthermore, earlier investigations have shown
that seed deposition in flooded areas is highly dependent on
flowpatterns andmicrotopography withinthe areas andthatthe
amountofseedsdepositedcoincideswiththedriftlineinflooded
areas(Nilsson&Grelsson,1990;Riis,Baattrup-Pedersen,Poulsen,
&Kronvang,2014).Wethereforeexpectedtofindthehighestdi-
versityatintermediatedistancefrom the streams.However,our
study didnotconfirmthisexpectationas thefunctionaldiversity
wasunaffectedbyflooding.Thisfindingindicatesthatspeciesar-
riving by water may not be more functionallydiversethanthose
arrivi ng by other means of di spersal. Th is interpretat ion is sup-
porte d by previous studie s reporting that species d ispersed by
hydrochor yareoftenthosealreadylocallyabundant(Brederveld,
Jähnig,Lorenz,Brunzel,&Soons,2011;Soomersetal.,2011)and
that floo ding in itself m ay not be sufficie nt to increase spe cies
richnes s in grasslan d vegetation u pon restor ation of more n atu-
ral floodingconditions(Baattrup-Pedersen,Riis, & Larsen, 2013;
Baattrup-Pedersen,Dalkvist,etal.,2013;Bissels,Holzel,Donath,
&Otte,2004).
FIGURE8 Averageresponseratios(±1SE)offunctional
traitdiversity(FDis)(a)andtraitcomposition(CWMs)(b)in
plotspositionedatthehighendofthefloodplain(position3).
Whenaresponseratioissignificantlydifferentfromzero,this
isindicatedwithanasteriskabovetheerrorbar(p<.05).Open
symbols(existing)comprisedataforthevegetationsurveys,
whereasclosedsymbols(seed)comprisedatafortheseedtrap
surveys.Thecoloroftheasteriskindicatesthetypeofvegetation
differingsignificantlyfromzero(i.e.,blackasterisk=seed,white
asterisk=existing).NotethatthescaleforFDisforBYC,LM,and
LSisdifferentincomparisonwiththeothertrait s
Existing Seed
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–6
–3
–9
9
6
3
0
–4
–2
–6
6
4
2
Drought Flooded
BYCSMSVCHLM
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–2
–1
–3
3
2
1
0
–6
–3
–9
9
3
1
0
–4
–2
–6
6
4
2
LS
DroughtFlooded
(a) Functional divergence (b) CWM
0
–2
–1
–4
–3
4
3
2
1
0
–2
–1
–4
–3
4
3
2
1
14
|
BAATTRUP- PEDERSEN E T Al.
5 | CONCLUSIONS
We observed large study site variability in plant community re-
sponses t o the hydrologic al conditio ns of our exper iment, rega rd-
ingbothdroughtandflooding.Wedid,however,identifyconsistent
patter ns in the taxo nomic and fu nctional r esponses of p lant com-
munitiestothealteredhydrologicalsettings.Bothtaxonomicdiver-
sity and functional diversity were generallynegativelyaffected by
floodingandtosomeextentalsobydrought.Thesefindingsindicate
that the range of successful strategies declined duetothe altered
hydrologic al setting s. The loss in fu nctional di versity was a lso re-
flecte d in the mean tra it response of th e riparian commu nity but
fewer signif icant and con sistent changes a ppeared in re sponse to
thealteredhydrologicalconditions.Thismightreflectacombination
ofthe existenceof severalstrategies withinthe vegetationto cope
withthealtered hydrological settings and adelayinthemeantrait
respons e due to a slow and pr ogressive fil ling of available n iches.
Takentogether,ourresultsdemonstratethateventhoughitisdiffi-
cultwithina3-yeartimeframetopredictgeneraleffectsofextreme
hydrologicalconditionsonriparianvegetationcharacteristicsacross
large regions, the observed losses in diversity likely affect ecosys-
tem func tioning by redu cing niche comp lementari ty with possi ble
cascadingeffectsonresourceuseefficiency.
ACKNOWLEDGMENTS
This work wa s supported by the Eu ropean Union 7th Fr amework
Projects RE FRES Hu nd ercontra ctno .24 4121andMARSunde rcon-
tract no.603378(Annet teBaattrup-Pedersen).Wethankthe land-
owners,waterboards, and nature organizationsforkindly allowing
ustousetheirareasfor ourexperiments,MarleneVenøSkjærbæk,
HenrikStenholt,UffeMensberg,andseveralstudentsforhardwork
constr ucting the dams , Helena Kalles trup, Tinna Chris tensen and
Juana Jacobsenforfigure layout,AnneMette Poulsenforeditorial
supportandFrederikHansenBaattrupformakingthefinaleditorial
changes.ThecompanyAstroTurf®isacknowledgedforsponsoring
seedmats.
CONFLICT OF INTEREST
Nonedeclared.
AUTHORS CONTRIBUTIONS
ABP,AG,CCH,andMSdesigned the study,EGandSELconducted
thestatistical analyses, AO and TR assistedinthefield campaigns,
andPMvD provideda number oftraitsforthespecies.ABPwrote
themanuscriptandallauthorscontributedtoitsfinalization.
ORCID
Annette Baattrup-Pedersen ht t p :// o rc i d .
org/0000-0002-3118-344X
REFERENCES
Ackerly, D.(2004).Functionalstrategiesofchaparralshrubs in relation
toseasonalwaterdeficitanddisturbance.Ecological Monographs,74,
25–44.ht tps://doi.org/10.1890/03-4 022
Alber t, C. H., Grassein, F., Schurr, F. M., Vieilledent, G., & Violle, C.
(2011). When and how should intraspecific variability be consid-
ered in trait-based plant ecology?. Perspectives in Plant Ecology,
Evolution and Systematics, 13, 217–225. https://doi.org/10.1016/j.
ppees.2011.04.003
Ander sen, H. E., K ronvang, B ., Larse n, S. E., Hof fmann, C . C., Jensen ,
T.S., & Rasmussen, E. K.(2006). Climate-change mediated impacts
onhydrologyandnutrient sina Danishlowland riverbasin. Science
of the Total Environment, 365, 223–237. https://doi.org/10.1016/j.
scitotenv.2006.02.036
Armstrong,W.,Brandle,R.,&Jackson,M.B.(1994).Mechanismsofflood
toleranceinplants.Acta Botanica Neerlandica,43,307–358.https://
doi .or g/10.1111/j.1438- 867 7.1994.t b0 0756.x
Baattrup-Pedersen, A., Dalkvist, D., Dybkjær, J. B., Riis, T., Larsen,
S. E., & Kronvang, B. (2013). Species recruitment following
flooding, sediment deposition and seed addition in restored ri-
parian areas. Restoration Ecology, 21, 399–408 . https://doi.
org /10.1111/j.1526 -100X .2012.0 0893 .x
Baattrup-Pedersen, A., Jensen, K., Thodsen, H., Andersen, H. E.,
Andersen, P. M., Larsen, S. E., … Kronvang, B. (2013). Ef fects of
strea m flooding on t he distribut ion and diver sity of ground water-
dependentvegetationinriparianareas.Freshwater Biology,58, 817–
827.ht tp s://doi.org/10.1111/fw b.12 08 8
Baat trup-Ped ersen,A .,Riis,T.,&Larsen,S.E.(2013).Catchmentcharac-
teristicsandplantrecruitmentfromsedimentinstreamandmeadow
habitats. River Research and Applications, 29, 855–863. https://doi.
org /10.1002/rr a. 2573
Bailey-Serres,J.,&Voesenek,L.A.C.J.(2008).Floodingstress:Acclimation
and geneticdiversity. Annual Review of Plant Biology,59, 313–339.
https://doi.org/10.1146/annurev.arplant.59.032607.092752
Banach,K.,Banach, A. M., Lamers, L. P.,DeKroon,H.,Bennicelli,R.P.,
Smits,A .J.,& Visser,E. J.(2009).Dif ferencesinflooding tolerance
betwe en species from t wo wetland habit ats with contra sting hy-
drology: Implications for vegetation development in futureflood-
water retention areas. An nals of Botany,10 3, 341–351.ht tps://doi.
org /10.1093/aob/m cn18 3
Barkman,J.J.,Doing,H.,&Segal,S.(1964).KritischeBemerkungenund
Vorschläge zur quantitativen Vegetationsanalyse. Acta Botanica
Neerlandica,13 ,394–419.https://doi.org/10.1111/j.1438-8677.1964.
tb0 0164.x
deBello,F.,&Mudr ak,O.(2013). Plant trait sas indic ators: Loss orgain
ofinformation?.Applied Vegetation Science,16,353–354.https://doi.
org /10.1111/avs c.1 2035
Beltman,B.,Willems,J.H.,&Güsewell,S.(20 07).Floodevent soverrule
fertilisereffec ts onbiomass production and speciesrichnessin riv-
erinegrasslands.Journal of Vegetation Science,18 ,625–634.https://
doi .or g/10.1111/j.1654-1103.2007.tb02576.x
Bernard-Verdier,M.,Navas, M. L.,Vellend, M., Violle, C.,Fayolle,A., &
Garnie r, E . (2012). Communit y assembly al ong a soil depth g radi-
ent:Contrastingpatternsofplanttraitconvergenceanddivergence
in aMediterranean rangeland. Journal of Ecolog y, 100, 14 22–1433.
https://doi. org/10.1111/1365-2745.120 03
Bissels , S., Holzel, N ., Donath, T. W., & Otte , A. (200 4). Evaluation o f
restor ation success in al luvial grassl ands under contr asting flood -
ing regimes. Biological Conservation, 118, 641–650. https://doi.
org /10.1016/j.biocon.200 3.10. 013
Borg y, B ., Violle, C., C holer, P., G arnier, E., Kattg e, J., Loranger, J., …
Diquelou, S. (2017). Sensitivity of community-level trait–environ-
mentrelationshipstodatarepresentativeness:Atestforfunctional
biogeography.Global Ecology and Biogeography,26,729–739.ht tps://
doi .or g/10.1111/geb.12573
|
15
BAATTR UP- PEDERSEN E T Al.
Braun-Blanquet, J. (1928). Pflanzensoziologie. Grundzüge der
Vegetationskunde.InW.Schoenichen(Hrsg.),Biol. Studienbücher 7
(pp.1–330).Berlin,Germany:Springer.
Breder veld, R. J., Jähnig, S. C., Lorenz, A . W., Brunzel, S., & Soons,
M. B. (2011). Dispersal as a limiting factor in the colonization
of restored mountain streams by plants and macroinverte-
brates. Journal of Applied Ecology, 48, 1241–1250. https://doi.
org/10.1111/j.1365-2664.2011.02026.x
van den Broek, T., van Dig gelen, R., & Bobbink, R. (2005). Variation
in seed buoyancy of species in wetland ecosystems with dif fer-
ent flooding dynamics. Journal of Vegetation Science, 16 , 579–586.
https://doi. org/10.1111/j .1654-1103 .2005.tb02399.x
Cappe rs, R., Be kker, R. M., & Jans, J . E. A. (200 6). Digitale Zadenatlas
van Nederland Groningen. Groningen, The Netherlands: Barkhuis
Publishing.www.zadenatlas.nl
Catford,J.A.,Naiman, R.J.,Chambers,L.E.,Roberts,J.,Douglas,M.,&
Davies, P.(2013). Predicting novel riparian ecosystems in a chang-
ing climate. Ecosystems, 16, 382–400. https://doi.org/10.1007/
s1002 1- 0 1 2-95 6 6-7
Chen, X., Huber,H.,deKroon,H.,Peeters,A.J., Poorter,H., Voesenek,
L. A., & Visser,E. J. (2009).Intraspecific variation inthemagnitude
and pat tern of floodin g-induced sh oot elongatio n in Rumex palus-
tris. Annals of Botany,10 4,1057–1067.https://doi.org /10.1093/aob/
mcp198
Christensen, J. H., & Christensen, O. B. (2007). A summar y of the
PRUDENCE model projections of changes in European climate
by the end of t his century. Climatic Change, 81, 7–30. http s://doi.
org /10.1007/s10584- 00 6-9210-7
Cornwell,W.K.,&Ackerly,D.D.(2009).Communityassemblyandshift s
inthe distribution of trait values across an environmental gradient
incoastalCalifornia.Ecological Monographs,79,109–126.http s://doi.
org /10.189 0/07-1134.1
Díaz, S.,&Cabido, M.(2001). Vive ladifference:Plantfunctional diver-
sitymatterstoecosystemprocesses.Trends in Ecolo gy & Evolution,16,
646–655.https://doi.org/10.1016/S0169-5347(01)02283-2
Diaz,S.,Hodgson,J.G.,Thompson,K.,Cabido,M.,Cornelissen,J.H.C .,
Jalili,A .,…Band,S.R.(2004).Theplanttraitsthatdriveecosystems:
Evidence fromthree continents. Journal of Vegetation Science, 15,
295–3 04.h tt ps: //doi.org /10.1111/j.1654 -1103.20 0 4.t b0 226 6.x
Douma, J.C., Bardin,V.,Bartholomeus,R .P.,& Bodegom,P.M.(2012).
Quantif ying thefunctional responses of vegetation todrought and
oxygen stress intemperateecosystems. – Func. Functional Ecology,
26,1355–1365.https://doi.org/10.1111/j.1365-2435.2012.02054.x
Do u vi lle ,H. ,C h auv in, F., Pla nto n,S ., Royer, J.F. ,S ala s- M el i a, D. ,&Ty tec a ,
S.(2002).Sensitivityofthehydrologicalcycletoincreasingamount s
ofgreenhousegasesandaerosols.Climate Dynamics,20,45–68.
Fraaije, R . G., Braa k, C. J., Verduy n, B., Bree man, L., Verh oeven, J. T.,
&Soons, M.B. (2015). Early plantrecruitment stages set the tem-
plate for the development of veget ation patterns along a hy-
drological gradient. Functional Ecology, 29, 971–980. https://doi.
org /10.1111/1365-2435.12441
Fraaije, R . G., Braa k, C. J., Verduy n, B., Verhoeve n, J. T., & Soons, M.
B. (2015). Dispersal versus environmental filtering in a dynamic
system : Drivers of vege tation pat terns and dive rsity alo ng stream
riparian gradients. Journal of Ecology, 103, 16 34–1646 . https ://d oi.
org /10.1111/1365-2745.1246 0
Garssen,A.G.,Baattrup-Pedersen,A .,Voesenek,L.A.,Verhoeven,J.T.,
& Soons, M . B. (2015). Rip arian plant co mmunity r esponses to i n-
creasedflooding:Ameta-analysis. Global Change Biology,21, 2 8 81–
2890.https://doi.org/10.1111/gcb.12921
Garssen, A. G ., Verhoeven, J. T., & Soons, M. B. (2014). Effects of
climate-inducedincreasesinsummer droughton riparianplantspe-
cies:Ameta-analysis.Freshwater Biology,59,1052–1063.https://doi.
org /10.1111/fwb.1 2328
Helsen, K., Hermy, M., & Honnay, O. (2012). Trait but not spe-
cies convergence during plant communit y assembly in restored
semi-natural grasslands. Oikos, 1 21, 2121–2130. https://doi.
org /10.1111/j.160 0-0706. 2012. 20 499.x
Horner,G.J.,Cunningham,S.C.,Thomson,J.R.,Baker,P.J.,&MacNally,
R.(2012).Foreststructure,floodingandgrazingpredictunderstorey
composi tion of flood plain fores ts in sout heastern A ustralia . Forest
Ecology and Management, 286, 148–158. https://doi.org/10.1016/j.
foreco.2012.08.023
Hough-Snee, N., Laub,B. G., Merritt,D. M.,Long, A.L., Nackley,L . L.,
Roper,B.B.,&Wheaton,J.M.(2015).Multi-scaleenvironmentalfil-
tersandnichepartitioninggovernthedistributionsofriparianvege-
tationguilds.Ecosphere,6,1–22.
Intergovernmental Panel on Climate Change (IPCC) (2007). Synthesis
report: Summary for policymakers.Fourth Assessment Report , IPCC
Plenar yXX VII,Valencia,Spain.
Jung, V., Albert, C . H., Violle, C., Kunstler, G., Loucougaray, G., &
Spiegelberger, T. (2014). Intraspecific trait variabilit y medi-
ates the response of subalpine grassland communities to ex-
treme dr ought events. Journal of Ecology, 102, 45–53. htt ps://doi.
org /10.1111/1365-2745.12177
Jung,V.,Violle,C.,Mondy,C.,Hoffmann,L.,&Muller,S.(2010).Intraspecific
variability and trait-based community assembly.Journal of Ecology,
98,113 4–1140 .https://doi.o rg /10.1111/j.1365-2745.2010.01687.x
Karlsson,I.B.,Sonnenborg,T.O.,Seaby,L.P.,Jensen,K.H.,&Refsgaard,
J.C .(2015). Effec tofa high-endCO2-emissionscenarioonhydrol-
og y. Climate Research,64,39–54.https://doi.org/10.3354/cr01265
Kleyer, M., & Bekker,R . M. (2008). The LEDAtraitbase: A database of
life-histor ytraitsoftheNorthwestEuropeanflora.Journal of Ecol ogy,
96,1266–1274.https://doi.org/10.1111/j.1365-2745.2008.01430.x
Kominoski, J. S., Shah, J. J. F.,Canhoto,C., Fischer,D.G., Giling, D. P.,
González, E., … McElarney, Y. R. (2013). Forecasting func tional
implications of global changes in riparian plant communities.
Frontiers in Ecology and the Environment, 11, 423–4 32. https://doi.
org /10.1890/120056
Lavorel, S., Grigulis, K., McIntyre, S., Williams, N. S., Garden, D.,
Dorrough,J.,…Bonis,A.(20 07).Assessingfunc tionaldiversityinthe
field-methodolog ymatters!.Functional Ecology,22,134–147.
Mommer, L.,De Kroon, H., Pierik, R., Bögemann, G. M.,& Visser,E. J.
(20 05).Af unction alc ompar isonofacc lim ationtoshad ean dsu bmer-
genceintwoterrestrialplantspecies.New Phytologist,167,197–206.
https://doi.org/10.1111/j.1469-8137.2005.01404.x
Nielsen , S. L., & San d-J ensen, K. (1989). Regul ation of photos ynthetic
rates of sub merged rooted m acrophytes . Oecologia, 81, 364–368.
htt ps://doi.org /10.10 07/BF0 0377085
Nilsson,C.,Brown,R.L.,Jansson,R.,&Merritt,D.M.(2010).Theroleof
hydrochoryinstructuringriparianandwetlandvegetation.Biological
Reviews,85,837–858.
Nilsson , C., & Grelsson , G. (19 90). The effec ts of litter displ acement
on riverbank vegetation.Canadian Journal of Botany, 68, 735–741 .
https://doi.org/10.1139/b90-097
Oddershede,A .,Svenning,J.C.,&Damgaard,C.(2015).Topographically
determined water availability shapes functional patterns of plant
communities within and across habitat types. Plant Ecology, 216,
1231–1242.https://doi.org/10.1007/s11258-015-0504-6
Oksanen,J.,Blanchet,F.G.,Kindt,R.,Legendre,P.,Minchin,P.R.,O’Hara,
R. B., … Wagner,H. (2014).Vegan:Communit y ecolog y package.R
package v ersion 2. 2-0. Retrieve d from http: //CR AN.R-pro ject.or g/
package=vegan.
Osenberg,C .W.,Sarnelle,O.,&Cooper,S.D.(1997).Effectsizeineco-
logica l experimen ts: The appl ication of bio logical mod els in meta-
analysis.American Naturalist,150,799–812.
Ozinga , W.A ., Bekker, R. M ., Schami nee, J. H., & Van G roenen dael, J.
M. (200 4). Dispersal potential in plant communities depends on
16
|
BAATTRUP- PEDERSEN E T Al.
environmentalconditions. Journal of Ecolog y, 92, 767–777.h ttps://
doi.org/10.1111/j.0022-0477.2004.00916.x
Poorter,L.,& Markesteijn,L.(2008). Seedling traitsdeterminedrought
tolerance of tropical treespecies. Biotropica,40,321–331.https://
doi .or g/10.1111/( ISSN)174 4-7429
RCoreTeam.(2014).R: A language and environment for statistical comput-
ing.Vienna,Austria:RFoundationforStatisticalComputing.http://
www.R-project.org/.
Riis, T., Baat trup-Peder sen, A., Poul sen, J. B., & Kronv ang, B. (2014).
Seedgerminationfromdepositedsedimentsduringhighwinterflow
in riparian areas. Ecological Engineering, 66, 103–110. https://doi.
org/10.1016/j.ecoleng.2013.05.006
vanRoosmalen,L., Sonnenborg, T.O.,&Jensen, K.H .(2009).Impactof
clim ateandla nd us ec ha ng eo nt he hydr ol og yofa la rge-sca le ag ri cul-
turalcatchment .Water Resources Research,45,W00A15.
Roscher, C., Schumacher, J., Gerighausen, U., & Schmid, B. (2014).
Different assemblyprocessesdrive shif ts in speciesand functional
compositioninexperimentalgrasslandsvaryinginsowndiversit yand
communityhistory.PLoS ONE,9,e101928. ht tps://doi.org/10.1371/
journal.pone.0101928
Sandel, B., Goldstein, L. J., Kraft, N. J., Okie, J. G., Shuldman, M. I.,
Ackerly, D. D., … Sud ing, K. N. ( 2010). Contras ting trai t response s
in plant communities to experimental and geographic varia-
tion in precipitation. New Phytologist, 188, 565–575. https://doi.
org /10.1111/j.1469- 8137.2 010.03 382.x
Soomer s, H., Sarneel, J. M.,Patberg,W.,Verbeek, S. K ., Verweij,P.A.,
Wassen, M . J., & van Diggele n, R. (2011). Factors in fluencing th e
seedsourceandsinkfunctions ofafloodplainnaturereserveinthe
Netherlands.Journal of Vegetation S cience,22,445–456.https://doi.
org /10.1111/j.1654-1103.20 11.01261.x
Ström, L.,Jansson, R., Nilsson, C., Johansson, M. E., &Xiong,S. (2011).
Hyd rolog iceffec tsonripar ianveget at ioninabo realR iver:Anexper-
imenttestingclimatechangepredictions.Global Change Biology,17,
254–267.https://doi.org/10.1111/j.1365-2486.2010.02230.x
Stromberg,J.C.,Bagstad,K. J.,Leenhouts, J.M.,Lite,S.J., &Makings, E.
(2005).Effectsofstreamflowintermittencyonriparianvegetationof
asemiaridregion river(San PedroRiver,Arizona).River Research and
Applications,21,925–938.https://doi.org/10.1002/(ISSN)1535-1467
Strombe rg, J. C., H azelton, A . F.,& Wh ite, M. S. (20 09). Plant sp ecies
richness in ephemeral and perennial reaches of a dryland river.
Biodiversity and Conservation,18,663–677.https://doi.org/10.1007/
s10531-0 08-9532-z
Suding,K.N.,Lavorel,S.,Chapin,F.S.,Cornelissen,J.H.,Diaz,S.,Garnier,
E.,…Navas,M.L.(2006).Scalingenvironmentalchangethroughthe
community-level:A traitbasedresponse andeffectframeworkfor
plants.Global Change Biology,14,1125–1140.
Swenson , N. G., & Enquist , B. J. (2007). Eco logical and evol utionary
determinants of a key plant functionaltrait: Wooddensity andits
community-widevariation acrosslatitudeand elevation. American
Journal o f Botany,94,451–459.https://doi.org/10.3732/ajb.94.3.451
Thodsen, H., Baattrup-Pedersen, A., Andersen, H. E.,Jensen, K. M.
B., Andersen, P. M., Bolding, K., & Ovesen, N. B. (2014). Climate
changeeffectsonlowlandstreamfloodregimesandriparianrich
fen vegetation communities in Denmark. Hydrological Sciences
Journal,51,1–15.
Thuille r,W., Al bert, C . H., Dub uis, A. , Randin, C ., & Guis an, A. (20 10).
Variation inhabitatsuitabilitydoesnotalwaysrelatetovariation in
specie s plant fun ctional t raits . Biology Letters,6, 120–123.https://
doi.org/10.1098/rsbl.2009.0669
vanderMaarel,E.(2007).Transformationofcover-abundancevaluesfor
appropriatenumericaltreatment– Alternativesto theproposalsby
Podani.Journal of Vegetation Science,18,767–770.
Violle, C., Bonis, A., Plantegenest, M., Cudennec, C., Damgaard, C.,
Marion , B., … Bouzillé , J. B. (2011). Plant fun ctional tr aits capt ure
species richness variations along a flooding gradient. Oikos, 120,
389–398.https://doi.org/10.1111/j.1600-0706.2010.18525.x
Violle,C., Navas,M.L.,Vile,D.,Kazakou,E.,For tunel,C.,Hummel,I.,&
Garnier,E.(2007).Lettheconceptoftraitbefunctional!.Oikos,116 ,
882–8 92. ht tp s://doi .org/10.1111/ j.0030 -1299.20 07.15559.x
Voesenek, L. A. C. J., Colmer, T. D., Pierik, R., Millenaar, F. F., &
Peeters, A. J. M. (2006). How plants cope with complete
submergence. New Phytologist, 170, 213–226. https://doi.
org /10.1111/j.1469- 8137.2 00 6. 01692.x
Voesenek,L.A.C.J.,Rijnders,J.H.G.M.,Peeters,A.J.M.,VandeSteeg,
H. M., & DeKroon,H. (2004). Planthormones regulate fastshoot
elongation underwater: From genes to communities. Ecology, 85,
16–27.https://doi.org/10.1890/02-740
Weiher, E., Cla rke, G. P., & Keddy, P. A . (1998). Communit y assembly
rules,morphologicaldispersion,andthecoexistenceofplantspecies.
Oikos,81,309–322.https://doi.org/10.2307/3547051
Westoby,M.,&Wright, I. J. (2006).Land-plant ecology on the basis of
functionaltraits.Trends in Ecolog y & Evolution, 21,261–268 .https://
doi.org/10.1016/j.tree.2006.02.004
Westwood, C. G., Teeuw, R. M., Wade, P. M., Holmes, N . T. H., &
Guyard,P.(2006).Influencesofenvironmentalconditionsonmac-
rophytecommunitiesindrought-affectedheadwaterstreams. River
Research and Applications, 22, 703–726. https://doi.org/10.1002/
(ISSN)1535-1467
Wright, I. J., Reich, P.B., Cornelissen,J.H .,Falster, D.S.,Groom,P.K.,
Hikosaka,K.,…Osada,N.(20 05).Modulationofleafeconomictraits
and tra it relations hips by climate . Global Ecolog y and Biogeography,
14,411–421.https://doi.org/10.1111/j.1466-822x.2005.0 0172.x
Wright,I.J.,Reich,P.B.,&Westoby,M.(2003).Least-costinputmixtures
ofwater and nitrogen for photosynthesis. American Naturalist, 161,
98–111.
Zwicke, M., Alessio, G . A., Thiery, L., Falcimagne, R., Baumont, R.,
Rossignol,N.,…Picon-Cochard,C.(2013).Lastingef fect sofclimate
disturbance on perennial grassland above-ground biomass pro-
duction under two cutting frequencies.Global Change Biology, 19,
3435–3448.
How to cite this article:Baattrup-PedersenA,GarssenA,
GötheE,etal.Structuralandfunctionalresponsesofplant
communitiestoclimatechange-mediatedalterationsinthe
hydrologyofriparianareasintemperateEurope.Ecol Evol.
201 8; 0 0 :1–16 . https://doi.org/10.1002/ece3.3973
