Vegetation structure and photosynthesis respond rapidly to restoration in young coastal fens
Full text
Vege a ion s uc u e and pho osyn hesis espond apidly o
es o a ion in young coas al ens
Anna M. Laine
1,†
, Anne Tol anen
2,3
, Lau i Meh €
a alo
4
& Ee a-S iina Tui ila
5
1
Depa men o Fo es Science, Uni e si y o Helsinki, P.O. Box 27, FI-00014 Helsinki, Finland
2
Depa men o Ecology, Uni e si y o Oulu, P.O. Box 3000, FI-90014 Oulu, Finland
3
Na u al Resou ces Ins i u e Finland (Luke), Uni e si y o Oulu, P.O. Box 413, FI-90014 Oulu, Finland
4
School o Compu ing, Uni e si y o Eas e n Finland, P.O. Box 111, FI-80101 Joensuu, Finland
5
School o Fo es Sciences, Uni e si y o Eas e n Finland, P.O. Box 111, FI-80101 Joensuu, Finland
Keywo ds
CO
2
, o es y d ainage, low esis ance,
pea land es o a ion, plan unc ional ype,
esilience, species composi ion, succession.
Co espondence
Anna M. Laine, Depa men o Fo es
Science, Uni e si y o Helsinki, P.O. Box 27,
FI-00014 Helsinki, Finland.
Tel: +358400826419;
Fax: +358 8 553 1061;
E-mail: [email p o ec ed]
P esen add ess
†
Depa men o Ecology, Uni e si y o Oulu,
P.O. Box 3000, FI-90014 Oulu, Finland
Funding In o ma ion
The Academy o Finland (G an /Awa d
Numbe : “287039”) Na u al Resou ces
Ins i u e Finland, Uni e si y o Helsinki.
Recei ed: 1 June 2016; Re ised: 29 June
2016; Accep ed: 30 June 2016
Ecology and E olu ion 2016; 6(19): 6880–
6891
doi: 10.1002/ece3.2348
Abs ac
Young coas al ens a e a e ecosys ems in he i s s ages o pea land succes-
sion. Thei d ainage comp omises hei successional de elopmen owa d
u u e ca bon (C) ese oi s. We p esen he i s s udy on he success o
hyd ological es o a ion o young ens. We ca ied ou ege a ion su eys a
six young ens ha ep esen und ained, d ained, and es o ed managemen
ca ego ies in he Finnish land upli coas be o e and a e es o a ion. We
measu ed plan le el ca bon dioxide (CO
2
) assimila ion and chlo ophyll luo-
escence (F /Fm) om 17 mos common plan species p esen a he si es.
Wi hin 5 yea s o es o a ion, he ege a ion composi ion o es o ed si es had
s a ed o mo e owa d he und ained baseline. The co e o sedges inc eased
he mos in esponse o es o a ion, while he co e o deciduous sh ubs
dec eased he mos . The apid esponse indica es high esilience and low esis-
ance o young en ecosys ems owa d changes in hyd ology. Fo bs had highe
pho osyn he ic and espi a ion a es han sedges, deciduous sh ubs, and
g asses, whe eas a es we e lowes o e e g een sh ubs and mosses. The
impac o managemen ca ego y on CO
2
assimila ion was an indi ec conse-
quence ha occu ed h ough changes in plan species composi ion: Inc ease
in sedge co e ollowing es o a ion also inc eased he po en ial pho osyn he ic
capaci y o he ecosys em. Syn hesis and applica ions. Res o a ion o o es y
d ained young ens is a p omising me hod o sa egua ding hem and b inging
back hei unc ion as C ese oi s. Howe e , hei low esis ance o wa e
able d aw down in oduces a isk ha egene a ion may be pa ially hinde ed
by he hea y d ainage in he su ounding landscape. The e o e, es o a ion
success is bes sa egua ded by managing he whole ca chmen s ins ead o
ca ying ou small-scale p ojec s.
In oduc ion
While p ima y succession has been a common o m o
pea land ini ia ion a e deglacia ion, no he n pea land
succession se ies a e cu en ly globally a e. They can
p esen ly be ound only in h ee egions wi h ongoing
land upli : a he coas s o Bo hnia Bay in Finland and
Sweden, coas o Whi e Sea in Russia, and in he Hud-
son Bay Lowlands o Canada. Young coas al ens a e he
i s s ages o p ima y pea land succession and occu in
loca ions whe e equen loods in luence he ege a ion
(Klinge and Sho 1996). In he beginning o he succes-
sion, la ge empo al a ia ion in wa e able is cha ac e -
is ic due o he shallow pea laye , low wa e holding
capaci y o he sandy soil, and lack o es ablished ege a-
ion ypical o la e successional s a es (Lepp€
al€
a e al.
2008; Tui ila e al. 2013). Wa e logged condi ions sup-
po accumula ion o o ganic ma e as pea and while
succession p oceeds, mi e hyd ology is dec easingly con-
olled by allogenic ac o s such as p ecipi a ion, su ace
in low and uno , and a ia ion in sea le el. This au o-
genic succession p oceeds om mine o ophic,
6880 ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d.
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.
g oundwa e - ed ens owa d omb o ophic bogs
(Hughes and Dumayne-Pea y 2002; Rydin and Jeglum
2013).
Du ing succession, esis ance inc eases and esilience
dec eases (Odum 1969). Resis an sys ems a e able o
main ain a gi en s a e when hey a e subjec ed o dis-
u bances. Resilien sys ems a e sensi i e o dis u bances,
ha is, hey espond by changing hei s uc u e and
unc ions, bu eco e apidly owa d hei o iginal s a e
when he sou ce o dis u bance is emo ed (Allison and
Ma iny 2008). P e ious s udies on he impac s o o -
es y d ainage and es o a ion show ha pea lands end
o con o m o he heo y o Odum (1969): The nu i-
en -poo , la e-successional bogs appea o be esis an
sys ems wi h slow and o en pa ial seconda y succes-
sion ollowing d ainage (Laine e al. 1995), and slow
egene a ion a e es o a ion (Jauhiainen e al. 2002;
Laine e al. 2011). In con as , young ens ha lack he
bu e ing hick pea laye ypical o bogs can be
expec ed o ha e low esis ance o hyd ological dis u -
bances.
In Finland, he occu ence o young coas al ens is
de ined o a zone <20 m a.s.l. (Rehell and Heikkil€
a 2009).
Thei succession s a s simila ly o, o example, Eu opean
dune slacks, whe e he ba e mine al soil is colonized by
pionee species ha a e adap ed o empo a ily o pe ma-
nen ly wa e logged condi ions (G oo jans e al. 1998).
Young coas al ens co e ed some 74,000 ha bu a e
s ongly impac ed by land managemen and app oxi-
ma ely 95% ha e been d ained ( ough es ima es based on
Rehell and Heikkil€
a 2009). The pe cen age is conside ably
highe han he a e age pea land d ainage pe cen age,
50%, o he o iginal a ea o 10.4 million hec a es in Fin-
land (Finnish S a is ical Yea book o Fo es y 2014).
In ensi e d ainage is he main eason ha pea land suc-
cession se ies a e conside ed c i ically endange ed (Rehell
and Heikkil€
a 2009).
Pea land d ainage o o es y, which impac s app oxi-
ma ely 15 million hec a es o pea lands globally (Paa i-
lainen and P€
ai €
anen 1995), al e s he hyd ological
egime, inc eases he ae a ion o pea , and edi ec s he
de elopmen owa d o es succession (Laine e al. 1995;
Vompe sky and Si in 1997; M€
alson e al. 2008). The
landscape di e si y dec eases due o he eplacemen o
pea land plan species by species common in he su -
ounding o es s (Vasande 1987). Changes in he species
composi ion may in luence ecosys em unc ions, such as
pho osyn hesis (e.g., Reich e al. 1998), biomass p oduc-
ion, and li e decomposabili y (Laiho 2006). Fo exam-
ple, inc eased shading by ees and lowe ed wa e able
may di ec ly in luence he a e o pho osyn hesis
(Med ano e al. 2003). The inc eased pea espi a ion
a e lowe s he C accumula ion capaci y (Minkkinen
e al. 2007) ha is a key ecosys em se ice o pea lands.
In young ens, d ainage p ohibi s he C accumula ion
ha would o he wise occu du ing he successional pea -
land de elopmen .
Res o a ion aims o assis he eco e y o he s uc u e
and unc ion o damaged ecosys ems owa d ha o
p is ine ecosys ems (e.g., Hobbs and C ame 2008). In
global and na ional policies, i is cu en ly seen as a c u-
cial means o sa egua d biodi e si y (Aichi Biodi e si y
Ta ge s 2011; EU Biodi e si y S a egy o 2020). The
p incipal es o a ion me hods o o es y d ained bo eal
pea lands a e he blocking o di ches o ec ea e he high
wa e able and he emo al o excess ees o educe
he anspi a ion a e and eins a e he landscape ypical
o na u al pea lands (e.g., Ta ainen e al. 2013). The
high wa e able can be quickly es o ed and common
pea land species, such as Sphagnum mosses and sedges
may apidly ecolonize he si es (e.g., Haapaleh o e al.
2011). Howe e , specialis -species, such as hepa ics, a e
ich en indica o s, and hollow/ la k species may s ill be
missing om es o ed si es se e al yea s a e es o a ion
(Hedbe g e al. 2012; Maana ilja e al. 2014). The
desi ed species and plan unc ional ype composi ion
depends on he pea land ype. Young ens a e ypically
domina ed by sedges, g asses, and o bs, and hei moss
laye is sca e ed (Lepp€
al€
a e al. 2008). Hence, high
amoun s o mosses o sh ubs a e no expec ed a e
es o a ion, as o bogs.
Because es o a ion o young ens has no been imple-
men ed o da e, li le in o ma ion exis s on he egene a-
ion o he s uc u e and unc ion o hese ecosys ems.
Knowledge gained om so-called ue pea lands wi h
hick pea laye s may no be applicable o young en
ecosys ems due o di e ences in pea hickness, plan spe-
cies composi ion, nu ien dynamics, and hyd ology,
which may c ucially in luence pea land esponses o d ai-
nage and es o a ion.
This is he i s s udy o quan i y he impac s o o -
es y d ainage and es o a ion on wa e able, ege a ion
composi ion, and unc ioning in e ms o CO
2
assimila-
ion in he a e young en ecosys ems. We hypo hesize
ha (1) due o he low esis ance o young ens o dis-
u bance, d ainage has changed hei hyd ology and
ecosys em unc ions, and di ec ed hei successional
pa hway owa d o es ; (2) as a consequence o he low
esis ance, ewe ing wi h elling edi ec s hei ege a ion
composi ion owa d ha o und ained ens al eady du -
ing he i s yea s a e es o a ion, ha is, a o s sedges,
g asses, and o bs; and (3) he expec ed change in ege-
a ion is e lec ed as inc eased pho osyn he ic a es while
sh ubs ha domina e in d ained pea lands a e eplaced
by g aminoids and o bs wi h highe pho osyn he ic
capaci y.
ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d. 6881
A. M. Laine e al. Impac s o D ainage and Rewe ing on Young Fens
Ma e ial and Me hods
S udy a ea
The s udy was ca ied ou in he Finnish land upli coas
o he Gul o Bo hnia in Siikajoki (64°480N, 24°380E).
The 30 yea s (1979–2009) a e age p ecipi a ion and mean
annual empe a u e a e 539 mm and 2.6°C, espec i ely,
and he leng h o g owing season is 150 days (Re onlah i,
Siikajoki, 64°410N, 25°050E, 48 m a.s.l., Finnish Me eo-
ological Ins i u e). We selec ed six coas al ens belonging
o h ee managemen ca ego ies in 2005: wo und ained
(UD1, UD2), wo d ained (D1, D2), and wo d ained
si es o be es o ed la e (R1, R2). The si es ep esen he
same p ed ainage coas al young en ype ha was o med
by p ima y mi e de elopmen ollowing pos glacial land
upli app oxima ely 100–200 yea s ago (Ekman 1996).
They a e loca ed in small dep essions o ~0.5–3ha
be ween nu ien -poo sand dunes. The si es lie a
1.5–2 m asl and ha e a 5- o 10-cm- hick o ganic soil
laye . The opog aphy o he a ea is di e se and undula -
ing. Finding exac ly simila s udy si es was he e o e no
possible.
The wo und ained si es a e loca ed in a small na u e
p o ec ion a ea ~4 km om he d ained and es o ed si es
ha a e loca ed app oxima ely 700 m apa . The plan
communi y o he und ained si es is composed o g ami-
noids and o bs. Moss laye is sca e ed and ba e soil
o ms a majo pa o he g ound (Table S1).
In d ained s a e, he g ound laye ege a ion was
mainly o med by sh ubs al hough a D2 and R2 sedges
we e also abundan . The ee s and olume o he wo
d ained si es was 71 and 21 m
3
(D1, D2, espec i ely),
and in es o a ion si es, i was 87 and 10 m
3
(R1, R2,
espec i ely). Res o a ion o he si es R1 and R2 was ca -
ied ou in wo phases in 2008. Fi s , be ween 75% and
100% o he s ems we e elled in sp ing, and 0–5 ees le
pe 100 m
2
. The ee s and densi ies co esponded o
hose be o e d ainage ha we e es ima ed using ae ial
pho os om he 1960s. Second, di ches we e blocked
using exca a o s in Augus 2008.
Wa e able and ege a ion composi ion
measu emen s
Wa e able (WT) was measu ed om pe o a ed plas ic
ubes, 3 cm in diame e , d illed in o 50 cm dep h in he
g ound. In each si e, he e we e 3–6 addi ional 100-cm-
long ubes ha we e used o co ec he measu emen s
du ing ex eme d ough s. In he d ained and es o ed
si es, WT measu emen s we e ca ied ou in May, Augus ,
and Oc obe om 2005 o 2014 and in he und ained
si es in 2007 and in 2010–2014. Vege a ion was eco ded
a 12–14 pe manen sample plo s o 1 91 m in he
es o ed and d ained si es 2 yea s be o e es o a ion in
2006 and 1 and 5 yea s a e es o a ion in 2009 and
2013, espec i ely. Fo he und ained si es, ege a ion was
eco ded a six 0.6 90.6 m sample plo s in 2007, 2010,
and 2013. The und ained si es we e pa o ano he
esea ch p ojec , which explains hei di e en sampling
p o ocol. The spa ial a ia ion o he ege a ion was low
a he und ained si es, which allowed hei lowe numbe
o sample plo s. Pe cen age co e s o plan s we e de ined
isually by consensus o wo expe ienced esea che s
wo king oge he , using he scale 0.25%, 0.5%, 1%, 2%,
3%, e c. We used quad a ames, di ided in o ou 25%
segmen s by s ings, o assis he obse a ion.
CO
2
assimila ion and chlo ophyll
luo escence measu emen s
To quan i y he ligh esponse o pho osyn hesis, CO
2
assimila ion (A) was measu ed in July 2011, wi h a po a-
ble open, ully con olled, low h ough gas exchange lu-
o escence measu emen sys em (GFS-3000; Walz,
Ge many) unde a ying ligh le els. To accoun o he
di ec e ec s o d ainage and es o a ion, we selec ed
common plan species occu ing in all si es so ha he
e ec on he same species can be compa ed be ween si es.
To s udy he indi ec e ec , ha is, e ec caused by
changing plan species composi ion due o managemen ,
we chose he mos common species wi hin each si e
(Table S2). Six samples pe species pe managemen ca e-
go y (und ained, d ained, and es o ed) we e measu ed.
Hal o he samples we e collec ed wi hin open- op cham-
be s (OTC) loca ed in each si e o a s udy o global
wa ming impac s. As he OTC ea men had no impac
on CO
2
assimila ion a es ( es ed wi h gene al linea
models, see below), hose samples we e included in his
s udy. The de ice was se up nea by he s udy si es, and
plan s we e picked a maximum 30 min be o e he mea-
su emen wi h an ample amoun o oo s and soil, hen
kep mois and in shaded condi ions. Depending on he
species, we enclosed one o se e al lea es wi hin he cu -
e e. A s anda d lea cu e e was used o all species o he
han sh ubs and mosses, o which he coni e cu e e
was used. CO
2
assimila ion was measu ed a 800, 50, 20,
and 0 lmol m
2
sec
1
pho osyn he ic pho on lux den-
si y (PPFD). The sample was allowed o adjus o cu e e
condi ions o 5 min be o e he i s measu emen and
hen o 3 min a e each change in he PPFD le el;
o he wise, he cu e e condi ions we e kep cons an
( empe a u e 20°C, CO
2
concen a ion 380 ppm, ela i e
humidi y 60%, low a e 400, and impelle in le el 5).
The ime equi ed o a ull measu emen cycle was
20 min.
6882 ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d.
Impac s o D ainage and Rewe ing on Young Fens A. M. Laine e al.
Each sample was s o ed in a pape bag and anspo ed
o he labo a o y, whe e hey we e s o ed in a cold oom
un il he d y weigh was measu ed a e d ying a 105°C
o a leas 12 h. We calcula ed CO
2
assimila ion pe d y
mass (lmol g DM
1
h
1
).
The in insic quan um e iciency o PSII (F /Fm) was
measu ed wi h a po able luo ome e (FMS-2; Hansa -
ech, King’s Lynn, UK) om in ac samples on si e.
This allowed measu emen o mo e species and a
highe sample size (Table S2). Lea es we e da k accli-
ma ed wi h lea clips (Hansa ech) o a leas 20 min
be o e he measu emen .
To quan i y he species speci ic chlo ophyll pigmen
concen a ions (Ensminge e al. 2001), six lea samples
om all plan species used in CO
2
assimila ion measu e-
men s we e collec ed and ozen in liquid ni ogen imme-
dia ely in ield, and s o ed a 80°C o 4 mon hs. P io
o analysis, lea samples (~50 mg) we e g ound in liquid
ni ogen, eeze d ied, and ex ac ed wi h 100% ace one
bu e ed wi h sodium bica bona e (NaHCO
3
) o 2 h a
4°C. Pigmen concen a ions in he ex ac s we e de e -
mined using a spec opho ome e (Shimadzu UV-1700,
Kyo o, Japan) in he labo a o y o he Depa men o
Biology, Uni e si y o Oulu.
Da a analysis
We applied de ended co espondence analysis (DCA) o
e eal he main communi y g adien s wi hin he da a se
and o isualize he empo al mo emen o communi ies
wi hin managemen ca ego ies o e he g adien s/o dina-
ion space. The empo al changes in di e en manage-
men ca ego ies we e compa ed using p incipal esponse
cu es (PRC). In he analysis, he measu ing yea s o
unmanaged si es we e conside ed simila o hose o
managed si es. This me hod is a de i a i e o edundancy
analysis (RDA) ha ocuses on he di e ences be ween
he species composi ions o he ea men s a each sam-
pling da e. I allows he ime ajec o y o species compo-
si ion in he con ol ea men o be displayed as a
ho izon al line agains which de ia ions in species compo-
si ion unde o he expe imen al ea men s can be plo -
ed. The abundance o each species is modeled as a sum
o h ee e ms: he species’ mean abundance in he con-
ol, a da e-speci ic ea men e ec , and an e o (Poulin
e al. 2013). The analyses we e pe o med wi h Canoco
5.01 o Windows.
To u he analyze he impac o he managemen ca e-
go y on ege a ion, plan species we e pooled o nine
plan unc ional ypes (PFT): sedges, g asses, o bs, e e -
g een sh ubs, deciduous sh ubs, Sphagnum, o es mosses,
mi e mosses, and li e wo s (Table S1). Fo each PFT, we
es ed he e ec o managemen , measu ing yea (2006/
2007, 2009/2010, 2013), hei in e ac ion, and WT on he
co e o he PFT wi h a linea mixed model. The ull
model o each PFT was
yijk ¼TiþYRkþTiYRkþw ijk þaiþbij þeijk (1)
whe e y
ijk
is he ans o med co e o he PFT in yea k
o sample plo jon si e i, T
i
is he ixed e ec o man-
agemen , YR
k
is he ixed e ec o yea k,T
i
*YR
k
, is he
in e ac ion o managemen and yea , and w
ijk
is he
ixed e ec o wa e able. Te ms a
i
and b
ij
a e no mally
dis ibu ed, ze o-mean andom e ec s o si e and plo
wi hin si e, and e
ijk
is he esidual e o . The ans o -
ma ion o y
ijk
was an a csin ans o ma ion o a powe -
ans o med co e , whe e he powe was de e mined o
each esponse so ha he model showed a cons an a i-
ance wi h no clea ends in he mean o esiduals.
Wa e able was emo ed om he models when i was
s a is ically insigni ican and nega i ely impac ed he
model pe o mance, using condi ional F- es s wi h c i i-
cal P alue o 0.05. Once he inal model was ound,
pos hoc condi ional - es s we e ca ied ou o compa e
he di e ences be ween managemen ca ego ies in 2006/
2007 and 2013 using bo h und ained and d ained as a
con ol.
To de e mine he e ec s o PFT, managemen , and
chlo ophyll con en on he ligh esponse pa ame e s o
ne pho osyn hesis, we applied a nonlinea mixed-e ec s
model wi h he hype bolic ligh sa u a ion cu e (e.g.,
Lappi and Oke -Blom 1992):
Aksi ¼Rks þPMAXks PPEDksi
aþPPFDksi
þeksi (2)
whe e he esponse A
ksi
is he obse ed ne pho osyn hesis
exp essed on a d y weigh basis, and he p edic o
PPFD
ksi
is he pho osyn he ic pho on lux densi y o
measu emen io sample son si e k. The pa ame e s o
be es ima ed a e espi a ion (R
ks
), pho osyn he ic capac-
i y, ha is, he maximum a e o ligh -sa u a ed g oss
pho osyn hesis (PMAXks ), and he maximum quan um
yield o CO
2
assimila ion (a), ha is, ligh use e iciency
a low ligh . The esidual (e
ksi
) is no mally dis ibu ed
wi h mean ze o and cons an a iance. In he ull model,
pa ame e s a,R
ks,
and PMAXks we e w i en as linea unc-
ions o ixed p edic o s PFT, managemen ca ego y and
chlo ophyll con en , and andom e ec s o nes ed le els
o si e and sample wi hin si e. These linea submodels
we e included in he ligh sa u a ion cu e (2), and all
coe icien s we e es ima ed in one s ep. Howe e , he ull
model was no es imable, and model i ing included i -
ing di e en models. A e an es imable model was
ound, es s on he di e en e ec s we e conduc ed using
app oxima e condi ional F- es s (P>0.05). Fo his
ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d. 6883
A. M. Laine e al. Impac s o D ainage and Rewe ing on Young Fens
analysis, we e-g ouped he ou moss PFTs in o wo
PFTs by including Sphagnum imb ia um in mi e mosses
(Table S2). Sedge PFT and und ained ca ego y we e used
as he s anda d wi h which o he PFTs and managemen
ca ego ies we e compa ed. Final submodels o he pho o-
syn hesis pa ame e s in equa ion 2 we e
Rks ¼PFTks þT ea ks þchl þaks (3)
PMAXks ¼PFTks þT ea ks þchl þbkþbks (4)
aks ¼PFTks þT ea ks þchl þcks (5)
whe e PFT
ks
, T ea
ks
, and chl a e e ec s o PFT (se en
le els), managemen ca ego y ( h ee le els), and chlo o-
phyll a+bcon en (con inuous), espec i ely. Te ms
a
ks
,b
ks
,c
ks
a e i a ia e no mal andom e ec s o sam-
pleson si e k, and b
k
is uni a ia e no mal andom
e ec o si e. The andom e ec s ake in o accoun
a iabili y in P
MAX
and a ha was no explained by
he ixed e ec s, and ensu es ha he conduc ed es s
ake in o accoun he lack o independence caused by
he g ouped s uc u e o he da a. Nes ed andom
e ec s we e o iginally included in all h ee models a
wo nes ed le els (si e and sample), bu only hose
shown o be signi ican we e included in he inal mod-
els. Fo sedges, o bs, and combined mosses, which
occu ed in all managemen ca ego ies, we es ed he
in e ac ion be ween PFT and managemen ea men o
he P
MAX
pa ame e .
To de e mine he e ec s o PFT (Table S2) and man-
agemen on F /Fm a io, we used a linea mixed-e ec s
model (1), using he i h powe o F /Fm as he esponse
a iable.
All models we e i ed, and he es s we e pe o med
using package nlme o he R so wa e, ollowing he p o-
cedu es o Pinhei o and Ba es (2000, Chap e 8).
Resul s
Wa e able
Wa e able was highly a iable bo h spa ially and empo-
ally (Fig. 1). Be o e es o a ion, he WT was on a e age
20 cm lowe a he d ained/ es o ed ca ego ies han a
und ained ca ego y. The di e ences be ween he wo si es
wi hin each managemen ca ego y we e high: Si es D1
and R1 we e d ie wi h on a e age 12 cm lowe WT
be o e es o a ion han si es D2 and R2, espec i ely. The
di e ence in WT be ween he d ie si e D1 and he we e
si e D2 was on a e age 19 cm du ing he whole s udy
pe iod. Du ing he sp ings be o e es o a ion, si es D2
and R2 had WT close o o a su ace, while in D1 and
R1 WT s ayed below he su ace. Following es o a ion,
he WT le els a si es R1 and R2 became simila o hose
in he und ained si es, and he a e age di e ence be ween
hem dec eased o 3 cm. Seasonal a ia ion was high: Si e
D2 and all und ained and es o ed si es expe ienced
sp ing and au umn loods. Du ing a e y d y pe iod in
2013, he WT d opped o an ex emely low le el a all
si es (Fig. 1).
Vege a ion composi ion
Fi y plan species we e eco ded in he six s udy si es.
Acco ding o DCA, he main ege a ion g adien was
ela ed o he d ainage succession whe e mesic we land
species we e eplaced by o es species (le o igh along
he i s DCA axis) (Fig. 2A). Two yea s be o e es o a-
ion in 2006, he ege a ion composi ion o he ou
d ained si es was a iable and di e ed om ha o he
und ained ca ego y (Fig. 2B and C). In addi ion, ege a-
ion he e ogenei y was much g ea e in he d ained and
es o ed ca ego ies han in he und ained ca ego y. Fi e
yea s a e es o a ion, he ege a ion composi ion o
many sample plo s had mo ed owa d he composi ion o
–45
–35
–25
–15
–5
5
15
25
Wa e able, cm
UD1
UD2
D1
D2
–75
–65
–55
08.2005
10.2005
05.2006
08.2006
10.2006
05.2007
08.2007
10.2007
05.2008
08.2008
10.2008
05.2009
08.2009
10.2009
05.2010
08.2010
10.2010
5.2011
8.2011
10.2011
5.2012
8.2012
10.2012
5.2013
8.2013
10.2013
Da e
R1
R2
Res o a ion
Figu e 1. A e age wa e able dep h (cm) in
he six s udy si es measu ed in May, Augus ,
and Oc obe in 2005–2013. Nega i e alues
indica e wa e able le el below he soil
su ace. UD deno es o und ained, D o
d ained, and R o es o ed si es. Res o a ion
by di ch blocking and clea cu ing was ca ied
ou in 2008. The dep h o he pea laye is
<10 cm; he e o e, mos o he a ia ion on
wa e able occu s in he mine al soil
subs a um.
6884 ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d.
Impac s o D ainage and Rewe ing on Young Fens A. M. Laine e al.
he und ained ca ego y (Fig. 2B and D). PRC analysis
con i med his esul : All managemen ca ego ies o igi-
nally di e ed om each o he , bu he es o ed ca ego y
was mo e simila o he und ained ca ego y and u he
app oached he und ained ca ego y wi hin 5 yea s o
es o a ion (Fig. 3). The ege a ion o he d ained ca e-
go y also mo ed owa d und ained ca ego y, especially a
egula ly looded si e D2, bu his change was mo e mod-
es han in es o ed ca ego y (Fig. 3). Fi e new plan spe-
cies we e eco ded a e es o a ion in 2013 a he
o iginally we e si e R2 (Table S1). These we e ypical
species o ens: Calamag os is pu pu ea,E iopho um
angus i olium, Callie gon co di olia, Sphagnum imb ia um,
and S. squa osum. Fou moss species had disappea ed:
S amine gon s amineum, Dic anum polyse um, Pohlia
nu ans, and Poly ichum s ic um. A si e R1 i e new spe-
cies ypical o ens: E. angus i olium, Ca ex os a a, Po en-
illa palus is, Callie gon co di olia, and Sphagnum ussowii
and h ee o es moss species: Dic anum uscescens, Poly-
ichum commune, and P ilidium cilia e we e obse ed
a e es o a ion, while none o he species had disap-
pea ed. Se en en species we e ound a he und ained
si es bu no in he es o ed si es: Ag os is canina, Alnus
glu inosa, Equise um lu ia ile, Lysimachia hy si lo a,
Figu e 2. De ended co espondence analysis
(DCA) o he ege a ion composi ion da a
2006 (be o e es o a ion) and 2013 (5 yea s
a e es o a ion). Eigen alues o he i s and
second axes a e 0.868 and 0.626, espec i ely.
The axes oge he explained 24% o he
a ia ion in he da a. (A) Species ha ha e a
leas 5% i on bo h axes (28 o 50 species);
(B) mo emen o s udy si e cen e s be ween
2006 and 2013, he leng h o he a ow
indica es he amoun o mo emen , (C)
en eloped sample plo s o each si e in 2006,
and (D) en eloped sample plo s o each si e in
2013. Species ull names a e gi en in Table S1.
ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d. 6885
A. M. Laine e al. Impac s o D ainage and Rewe ing on Young Fens
Peucedanum palus e,Sphagnum allax, and S. subsecun-
dum (Table S1).
Impac o managemen on plan unc ional
ypes
We ound a signi ican managemen impac on sedges
and g asses (P<0.05, Table S3). P io o es o a ion, he
co e o deciduous sh ubs was highe and he co e o
g asses lowe s in he es o ed ca ego y han in he
und ained ca ego y. The co e o g asses and b own
mosses was lowe in he d ained ca ego y han in he
und ained ca ego y. Possibly due o low numbe o epli-
ca es, he e we e no s a is ically signi ican di e ences
be ween he es o ed and d ained ca ego ies (Table S4),
al hough he co e o sh ubs seemed o be g ea e , and
ha o o es mosses and Sphagnum lowe in he es o ed
ca ego y (Fig. 4).
The co e o PFTs di e ed be ween he h ee s udy
yea s, wi h he excep ion o he e e g een sh ubs, o es
mosses, and Sphagnum, which did no show any change
(Fig. 4, Table S3). The di ec ion o change, ha is, he
inc ease o dec ease in co e , was di e en be ween
he managemen ca ego ies (Fig. 4), which was seen as
he signi ican in e ac ion be ween managemen and yea
o all o he PFTs han o es mosses and Sphagnum.
(Table S3). Fi e yea s a e es o a ion, he sedge co e o
he es o ed ca ego y had inc eased o a highe le el han
ha a he und ained and d ained ca ego ies (Fig. 4).
Deciduous sh ub co e had dec eased a he es o ed ca e-
go y and was no longe signi ican ly highe han a he
und ained ca ego y (Table S4). A e inclusion o manage-
men ca ego y, he impac o WT was no signi ican o
he co e o any PFT (Table S3).
CO
2
exchange o plan unc ional ypes
Maximal pho osyn hesis (P
MAX
) a ied be ween PFTs.
Fo bs had signi ican ly highe , and e e g een sh ubs, mi e
mosses, and o es mosses lowe P
MAX
han had sedges
(Fig. 5, Table S5 Model 1, Table S6). Sedges we e chosen
as he s anda d o he models as hey occu in all s udy
si es. Respi a ion a e (R) a ied be ween PFTs and man-
agemen ca ego ies (Table S5). Fo bs had signi ican ly
highe and mosses signi ican ly lowe R alues han sedges
(Fig. 5, Table S6), and R was highe a he es o ed and
d ained ca ego ies han a he und ained ca ego y. The
maximum quan um yield o CO
2
assimila ion (a)
depended on PFT (Table S5). I was highe o o bs,
deciduous, and e e g een sh ubs han o sedges, and
lowe o mi e mosses han o sedges (Fig. 5, Table S6).
Chlo ophyll a+bcon en , which a ied be ween PFTs
(Table S2), had a signi ican e ec only o R (Table S5,
Model 2).
A e es ing he whole ange o PFTs ac oss he s udy
a eas, we ocused on he di ec e ec s o managemen on
sedges, o bs and combined o es and mi e mosses,
which occu ed in all managemen ca ego ies. All es i-
ma ed pa ame e s om Equa ion 2 (P
MAX
,R, and a) we e
signi ican ly a ec ed by PFT and managemen ca ego y so
ha ha plan s g owing on d ained ca ego y had signi i-
can ly highe P
MAX
alues han hose on und ained ca e-
go y (Model 3 in Tables S5 and S6). Simila ly o Model
1, o bs had highe and mosses lowe P
MAX
alues com-
pa ed o sedges (Table S6). In addi ion, we ound an
in e ac ion be ween managemen ca ego y and PFT
(Table S6), so ha mosses on d ained ca ego y had signi -
ican ly lowe P
MAX
alues han was gene ally ypical o
mosses. Simila ly o Model 1, espi a ion a e (R) was
613
PRC.Time
–2 10
PRC.1
UD1
UD2
R2
D2
R1
D1
–
34
Resp.1
Ag sCani
Ca xCans
Ca xNig
Ca xRos
Emp Nig
EquiFlu E ioAngs
GaliPals
LedmPals
LysmThy
My cGale
PeucPals
Ph gAus
PinsSyl
Po nPals
SalxPhyl
VaccUlig
VaccVi s
AulcPals
Dic Pols
PleuSch
Pol Comm
P ilPulc
ScapPald
SphgSubs
Wa nsSp
Figu e 3. P incipal esponse cu e (PRC)
analysis, wi h und ained si e 1 (UD1) as
e e ence; e ec o managemen /si e,
including i s in e ac ion wi h ime, is signi ican
acco ding o Mon e Ca lo pe mu a ion es
(F=33.4, P=0.002). UD =und ained,
D=d ained, and R = es o ed. Dashed e ical
line indica es he yea o es o a ion ( ime
poin 0). The line on he igh shows he
species sco es o he i s o dina ion axis. Fen
species om und ained si es a e a he bo om
end. Fo es species a e a he op.
6886 ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d.
Impac s o D ainage and Rewe ing on Young Fens A. M. Laine e al.
highe (i.e., mo e nega i e) a bo h es o ed and d ained
ca ego ies han a und ained ca ego y (Table S6, Model
3), and o bs had highe and mosses lowe espi a ion
a es han sedges. The maximum quan um yield o CO
2
assimila ion (a) was highe in bo h he es o ed and
d ained ca ego ies (P alue =0.063) han in he
und ained ca ego y, and o bs had highe and combined
mosses lowe a han sedges (P alue =0.081) (Table S6,
Model 3).
The in insic quan um e iciency o PSII (F /
Fm)
The measu ed F /Fm a io a ied be ween 0.21 and 0.87.
Sh ubs and g asses had signi ican ly highe and mi e
mosses signi ican ly lowe F /Fm a ios han sedges
(Fig. 6, Table S7), whe eas he e we e no di e ences in
F /Fm be ween managemen ca ego ies.
Discussion
The allogenic con ol o hyd ology and succession in
young coas al ens is somewha compa able o cen al
Eu opean dune slacks. In dune slacks, in e annual a ia-
ion in he wa e able, and especially he win e looding,
a e he mos impo an ac o s de e mining ege a ion
succession (G oo jans e al. 1998). This phenomenon can
also be obse ed in young coas al ens. Unlike in dune
slacks howe e , du ing mi e de elopmen , he au ogenic
con ol o wa e able gains a le el ha is able o suppo
pea o ming ege a ion and allows mi e succession o
p oceed a e he connec ion o he looding sea is los
(Tui ila e al. 2007; Lepp€
al€
a e al. 2011).
As he landscape whe e young coas al ens a e loca ed in
Finland is in ensi ely d ained, hei succession is la gely
dis u bed. The e is no ea lie in o ma ion on how hese
ecosys ems esponse o en i onmen al pe u ba ions such
0
10
20
30
40
50
Co e %
Sedges G asses Fo bs
0
10
20
30
40
50
Co e %
E e g een sh ubs Deciduous sh ubs Fo es moss
UD_6 UD_13 D_6 D_13 R_6 R_13
Sphagnum
0
10
20
30
40
50
Co e %
Mi e moss Li e wo s
Figu e 4. Co e SE o PFTs in managemen ca ego ies (UD =und ained, D =d ained, and R = es o ed) be o e es o a ion in 2006 and
5 yea s a e es o a ion in 2013.
ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d. 6887
A. M. Laine e al. Impac s o D ainage and Rewe ing on Young Fens
as d aining and es o a ion. Al hough coas al ens a e spe-
cial ecosys ems, he knowledge a ained om hem can be
used o assess he ou come o en i onmen al pe u ba ions
in analogous, allogenically con olled ecosys ems. They
se e as ecological labo a o ies whe e he ou come o en i-
onmen al changes can be es ed and apidly seen.
D ainage impac s in young ens
As p oposed in ou i s hypo hesis, he hyd ology, ecosys-
em unc ions, and ege a ion composi ion had changed as
a consequence o d ainage. The change ha was obse ed
in he successional pa hway wi hin h ee decades is gene -
ally obse ed also on olde o es y d ained pea lands bu
o e longe ime scale (Laine e al. 1995). The ela i ely
apid change indica es ha young ens ha e low esis ance
o dis u bances. Howe e , because hese ecosys ems a e
loca ed in lood-p one loca ions close o he sea and do no
ha e bu e ing pea deposi s, d ainage e iciency a ied wi h
local opog aphy. This esul ed in o a ange o
Sedge
Fo b
G ass
E e g een
sh ub
Deciduous
sh ub
Mi e moss
Fo es moss
–50
50
150
250
350
450
550
–50
50
150
250
350
450
550
A, µmol g(DM)–1 h–1
750 800 850
–50
50
150
250
350
450
550
–25 25 75
PPFD, µmol m–2 s–1
(A)
(B)
(C)
Figu e 5. A e age ca bon dioxide (CO
2
)
assimila ion o di e en plan unc ional ypes
measu ed a pho osyn he ic pho on lux
densi y (PPFD) 0, 20, 50, and
800 lmol m
2
sec
1
in (A) und ained, (B)
d ained, and (C) es o ed ca ego ies. B eakage
o x-axis be ween PPFD 75 and 750.
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
F /Fm (SE)
Und ained D ained Res o ed
Figu e 6. The in insic quan um e iciency o PSII (F /Fm) measu ed
o se en PFTs managemen ypes: und ained, d ained, and es o ed.
6888 ª2016 The Au ho s. Ecology and E olu ion published by John Wiley & Sons L d.
Impac s o D ainage and Rewe ing on Young Fens A. M. Laine e al.