RESEARCH/REVIEW ARTICLE
Impac s o win e eeding o eindee on ege a ion and soil in
he sub-A c ic: insigh s om a eeding expe imen
Minna Tu unen,
1
Pi i a Oksanen,
1
Te hi Vuojala-Magga,
1
Inke i Ma kkula,
1
Ma ja-Liisa Su inen
2
& Juha Hy o
¨nen
2
1
A c ic Cen e, Uni e si y o Lapland, P.O. Box 122, FI-96101 Ro aniemi, Finland
2
Finnish Fo es Resea ch Ins i u e, Ro aniemi Resea ch Uni , P.O. Box 16, FI-96301 Ro aniemi, Finland
Keywo ds
Reindee he ding; supplemen a y win e
eeding; sub-xe ic hea h o es ; o age.
Co espondence
Minna Tu unen, A c ic Cen e, Uni e si y o
Lapland, P.O. Box 122, FI-96101 Ro aniemi,
Finland. E-mail: [email p o ec ed]
Abs ac
The o e all impac s o win e eeding, including he e ec s o bo h o age and
eindee (Rangi e a andus), we e s udied in an expe imen in he Hammas-
un u i he ding coope a i e (688N), Finland, wi h 300 eindee in a sub-xe ic
hea h o es (35 ha) du ing Ma ch/Ap il 2009 and 2010. The eeding p ac ices
on he 50 plo s we e: (1) eeding wi h g ass silagehay wi h le o e s clea ed
in he sp ing; (2) eeding wi h g ass silagehay wi h le o e s no clea ed; and
(3) eeding wi h pelle s. In addi ion, (4) un enced and (5) enced con ol plo s
we e included, on which he eindee we e no ed. No in asi e plan species
in oduced h ough g ass silagehay we e obse ed on he plo s. The co e age
o Deschampsia lexuosa inc eased on g ass silagehay plo s a e he i s
win e , and bo h co e age and heigh o he g ass inc eased a e he second
win e . The co e age o Dic anum sp. and Pleu ozium sch ebe i was lowe on
plo s whe e eindee we e ed wi h g ass silage (le o e silage no clea ed) s.
con ols. Some dwa sh ubs, e.g., Calluna ulga is, showed a simila esponse
a e he second win e . The N con en o Empe um nig um was highe on g ass
silage plo s (le o e silage no clea ed) a e he i s win e s. con ols. A e
wo win e s, he N con en o bo h E. nig um and Vaccinium my illus was
inc eased. O he soil a iables s udied, C/N dec eased on g ass silagehay
plo s s. enced con ols. We conclude ha e en a mode a e le el o eeding
may cause changes ha can lead o a g adual shi om nu ien -poo hea h
o es owa ds a mo e nu ien - ich ype.
In no he n Fennoscandia and no h-wes Russia, ein-
dee he ding has been a ec ed in ecen decades by such
socio-economic changes as mechaniza ion, supplemen-
a y win e eeding, egula ions ega ding mea p oduc-
ion and pa asi e medica ion (Je nsle en & Kloko 2002;
Fo bes e al. 2006; Tyle e al. 2007). In addi ion, he
quali y and ex en o pas u e ha e declined conside ably
owing o, among o he ac o s, compe ing o ms o land
use, such as o es y, indus ial de elopmen , cons uc-
ion, mining and ou ism, as well as o e g azing caused
by he inc eased numbe o animals (Kumpula e al.
2009; Ma ila & Mikkola 2009). Fu he mo e, p essu es
exis o inc ease supplemen a y win e eeding: highe
eindee densi ies and inadequa e pas u e o a ion sys-
ems ha e impai ed he quali y o win e pas u es, and
he p edic ed luc ua ions in win e wea he and snow
condi ions may well educe he a ailabili y o na u-
al o age (Pu konen & Roe 2003; Nieminen 2010;
Callaghan e al. 2011; Hansen e al. 2011).
A changing clima e can ha e bo h posi i e and
nega i e impac s on he a ailabili y and quali y o
eindee o age (Weladji & Holland 2006; Tyle e al.
(page numbe no o ci a ion pu pose)
To access he supplemen a y ma e ial o his a icle, please see he supple-
men a y iles unde A icle Tools online.
Pola Resea ch 2013. #2013 M. Tu unen e al. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion-Noncomme cial 3.0
Unpo ed License (h p://c ea i ecommons.o g/licenses/by-nc/3.0/), pe mi ing all non-comme cial 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.
1
Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610
2007; Moen 2008; Tu unen e al. 2009; Tømme ik e al.
2012). I has been p edic ed ha he no he n la i udes
will become wa me and we e , especially du ing
au umn, win e and sp ing. Inc easing win e empe a-
u es and empe a u e luc ua ions will lead o mo e
equen snowmel , ain-on-snow e en s and eeze*
haw cycles, which in u n causes compac ion o he
snow pack, o ma ion o ice laye s abo e and wi hin he
snow co e , and ice encasemen (Pu konen & Roe 2003;
Solomon e al. 2007; Callaghan e al. 2011). Icing and
a ia ions in he hickness o he snow co e may wo sen
he condi ions unde which eindee dig o o age
and he eby inc ease hei ene gy expendi u e. In ac ,
dec eased a ailabili y o win e o age may be one eason
o he de e io a ion obse ed in he condi ion o animals
and o low cal ing pe cen ages (Helle & Kojola 2008;
Hansen e al. 2011; Vuojala-Magga e al. 2011).
In Finland, supplemen a y win e eed consis s p i-
ma ily o hay, g ass silage, g ain-based comme cial pelle s
o a combina ion o hese. Du ing he pe iod 200708, 18
million kg o pelle s and o e 40 million kg o hayg ass
silage (calcula ed as d y hay) we e used o eeding
(Nieminen 2010). In he cen al and no he n pa s o
he eindee he ding a ea, ield eeding is mo e common
han pen eeding, wi h mo e han 60% o all animals
being ed in he ield. Field eeding is always based pa ly
on na u al pas u es; in pens eindee ge all o a leas
mos o he o age om he he de . Supplemen a y
eeding has made i possible o keep he numbe and
condi ion o eindee s able despi e di icul win e
wea he and snow condi ions (A
˚hman & Danell 2006;
Nieminen 2010). A emale eindee in good condi ion
a e he win e will no mally gi e bi h o a heal hy cal
and be able o ake ca e o i well (Elo an a & Nieminen
1986). The cal es o eindee ed wi h supplemen a y
o age a e hea ie han hose o eindee ha o age only
on na u al pas u es (Sa
¨kkinen e al. 1999; Maijala &
Nieminen 2001). Ye , win e eeding may inc ease he
isk o diseases (Oksanen e al. 1997; T yland e al. 2001;
Asch alk e al. 2003). I may also in ensi y he widely
epo ed di ec impac s o eindee g azing on and
ampling win e pas u e ege a ion* educed lichen
co e in pa icula *as well as he indi ec impac s on
soil p ocesses, o example, C and N mine aliza ion and
o ganic ma e decomposi ion, which g ea ly depend on
he si e ype and g azing in ensi y (Va
¨ e e al. 1995;
Olo sson e al. 2001; Olo sson & Oksanen 2002, 2005;
Olo sson e al. 2004; Pajunen e al. 2008; Moen e al.
2009; Kumpula e al. 2011).
To ou knowledge, he ecological impac s o win e
eeding o eindee on ege a ion and soil ha e no been
s udied ea lie . The aim o his in es iga ion was he e-
o e o s udy he o e all impac s o supplemen a y win e
eeding on ege a ion and soil in a sub-xe ic hea h ype
o es in no he n Finland (688N). Ou main ocus was o
s udy he po en ial changes in species co e age in he
ield- and g ound-laye ege a ion as well as he N
con en o dwa sh ubs and soil. In his s udy, ‘‘ he
o e all impac s o win e eeding’’ include he e ec o
bo h o age (g ass silage, hay, pelle s) and eindee
ac i i y in he o m o g azing, ampling, digging and
d oppings. We p oceeded on he basis o h ee hypo h-
eses. Fi s , win e eeding causes changes in he species
co e age o he g ound and ield laye o he sub-xe ic
hea h o es . Secondly, win e eeding esul s in inc eased
N con en in he dwa sh ubs Vaccinium my illus and
Empe um nig um and in he N con en and pH o soil
below he humus laye . And hi dly, pelle eeding causes
ewe signi ican e ec s on ege a ion and soil han
eeding wi h g ass silagehay and ha g ass silage
hay eeding causes ewe e ec s on plo s whe e unea en
o age was clea ed, when compa ed o unclea ed plo s.
We we e also in e es ed in de e mining when he o e all
impac s o win e eeding on ege a ion and soil become
isible. The esea ch u ilized an ex ensi e con olled ield
expe imen simula ing win e eeding o eindee o e
wo win e s (200810) in he Hammas un u i he ding
coope a i e in Finland. The s udy was planned and
conduc ed in coope a ion wi h he local he de s.
Ma e ial and me hods
S udy si e
The expe imen was conduc ed app oxima ely 15 km
om he illage o Ku u a. The illage is loca ed in he
Hammas un u i he ding coope a i e (68821?N, 268
50?E), one o 13 Sami eindee he ding coope a i es in
no he n Finland (Fig. 1). The a ea is cha ac e ized by
bo eal coni e ous o es s, sub-A c ic moun ain bi ch
woodlands, mi es and ells (Oksanen & Vi anen 1995).
The 35 ha si e was selec ed o his s udy in coope a ion
wi h he local he de s. I has a ‘‘cold, snowy, o es
clima e’’ (Ko
¨ppen clima e ype: D c) ha is cha ac e ized
by high annual a iabili y. The mean annual empe a-
u e, measu ed om 1970 o 2000, was 0.618C, wi h
a mean empe a u e in July o 14.18C and in Janua y o
13.08C. Annual p ecipi a ion was 426 mm; abou hal
o his is snow, which accumula es o a dep h o 65 cm
and mel s on a e age by 17 May (da a ob ained om he
Finnish Me eo ological Ins i u e).
The la ges pe missible numbe o eindee in he
coope a i e is 5500 animals. The e a e 2.5 eindee /
km
2
. Win e pas u es in Hammas un u i con ain 12.2 ha
Impac s o win e eeding o eindee on ege a ion and soil M. Tu unen e al.
2
(page numbe no o ci a ion pu pose) Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610
o e icolous lichen pas u e/ eindee and 16.0 ha o
a bo eal lichen pas u e/ eindee . Some o he o es s in
he coope a i e a e p o ec ed, and some a e used o
comme cial o es y. The o al co e age o in as uc u e
is 0.8% (17 km
2
), and i s impac a ea is 6.7% (146 km
2
)
o he land a ea o he coope a i e (Kumpula e al. 2009).
Since he middle o he 1960s, he cal ing pe cen age
(numbe o cal es bo n pe 100 emales as calcula ed
du ing he summe o au umn cal ma kings) has anged
be ween 6 and 75%. I eached i s minimum in 1973
(6%) and 1974 (18%) due o a sha p decline in he
a ailabili y o win e o age, caused by di icul snow
condi ions. In he 2000s, he cal ing pe cen age anged
be ween 44 and 60% (Reindee He de s’ Associa ion
2011; Vuojala-Magga e al. 2011).
Win e he ding in Hammas un u i en ails he ding and
eeding in he ield o in cal ing ences. Hay, g ass silage o
pelle s a e used as eed in win e p incipally o mo ing
eindee om one place o ano he and keeping he he d
oge he . Cal ing ences, which acili a e sa e cal ing, a e
used om Ma ch/Ap il o he end o May, depending on
he annual snow condi ions. He de s began using he
ences in he mid-1990s because many cal es we e bo n
on snow o in wa e holes due o la e snowmel in he
sp ing. In ence cal ing, emale eindee a e collec ed and
kep in la ge enced o es a eas, which allows he de s o
ensu e ha he animals a e i and well. Towa ds he end o
he 2000s, snow began mel ing ea lie and some o he
siidas (ex ended amily o kin g oup consis ing o econom-
ically independen households) ha e changed om he
cal ing ence sys em o he sp ing he ding sys em, in
which eindee a e ed in he o es i snow is deep o ha d
(Vuojala-Magga e al. 2011).
Feeding expe imen
The eeding expe imen , which simula ed supplemen a y
win e eeding o eindee , was conduc ed om 2 Ma ch
o 19 Ap il 2009 (49 days) and om 1 Ma ch o 30 Ap il
2010 (61 days). The expe imen al si e (35 ha) is a poo ly
egene a ed pine-domina ed sub-xe ic hea h o es
(Empe umMy illus ype [EMT]), wi h Sco s pine seed
ees le a e cu ings in he 19501960s. I is a ypical
o es y and win e pas u e a ea in he egion. The
c i e ia o selec ing he si e we e ha i had o p o ide
easy access by snowmobile o he de s in win e o daily
eeding and by ca o esea che s in summe o
conduc ing he ield measu emen s and sampling.
In June 2008, 50 expe imen al plo s (Ø 5 m, 19.6 m
2
)
we e es ablished on he expe imen al si e (35 ha), and
each plo was alloca ed andomly o one o he ollowing
eeding p ac ices: (1) eeding wi h g ass silagehay wi h
le o e s clea ed om he plo in he sp ing; (2) eeding
wi h g ass silagehay wi h le o e s no clea ed; and (3)
eeding wi h pelle s. The expe imen also included (4)
un enced and (5) enced con ol plo s, on which he
eindee we e no gi en supplemen a y o age. The
ences we e app oxima ely 2 m high and made o wi e
mesh. Reindee had no access o he enced plo s. Each
eeding p ac ice was eplica ed 10 imes ( o al numbe o
plo s: 50). The plo s we e a anged o o m an o al; he
dis ance be ween he closes plo s was 30 m and 1 km
be ween he a hes .
Fig. 1 Loca ion o he eindee win e eeding expe imen in he
Hammas un u i he ding coope a i e (ci cle) and o he wea he s a ion
in Saa iselka
¨(s a ). In he uppe map, he eindee he ding a ea in
Finland is indica ed in black.
M. Tu unen e al. Impac s o win e eeding o eindee on ege a ion and soil
Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610 3
(page numbe no o ci a ion pu pose)
Ou app oach was p ac ical a he han heo e ical,
which is why no enced plo s wi h pelle s o g ass
silagehay we e included in he expe imen ( eindee
he de s ed he eindee du ing he expe imen ). As all o
he eeding p ac ices in he expe imen (excep enced
and un enced con ol plo s) encompassed no only he
impac o o age (g ass silagehay, pelle s), bu also he
e ec o eindee on he pas u es in he o m o g azing,
ampling, digging and d oppings, we s udied he o e all
e ec o win e eeding on a ious ege a ion and soil
a iables. Ano he eason o his app oach was ou
expec a ion ha he e ec s o eindee on pas u e would
no be equal in he case o all o he eeding p ac ices. Fo
example, he a ac i eness o di e en ypes o o age o
eindee a ies, wi h he animals p e e ing pelle s o
g ass silagehay.
The o age used in he expe imen was a mix u e o
g ass silage p ese ed wi h molasses sy up and hay
(he ea e ‘‘g ass silagehay’’); i was p oduced 50100
km om he si e. The main o age species we e Fes uca
p a ensis Hudson, Deschampsia cespi osa (L.) Beau ., Phleum
p a ense L. and Ta axacum o icinale. The o age also
con ained T i olium epens L., Rumex ace osella L., Ranun-
culus ac is L., S ella ia media (L.) Vill., Raphanus aphanis-
um L., Achillea mille olium L., Rumex ace osa L., Ve onica
chamaed ys L. and Epilobium angus i olium L. The he de s
ga e eindee oughly 3 kg o o age/plo /day, which
equals 150 g o age/m
2
/day. The animals we e also ed
wi h comme cial eindee pelle s (Ta
¨h i-Po o A ic and
Ta
¨h i-Po o 2 Balans, Kinnusen Mylly L d., U aja
¨ i,
Finland), which con ained oa , whea , ba ley and molas-
ses. The he de s ga e eindee oughly 1 kg o pelle s/
plo /day, which equals 50 g pelle s/m
2
/day. The esul s o
he o age and pelle analyses a e p esen ed in Table 1.
Acco ding o he eindee he de s pa icipa ing in his
s udy, he eindee dug app oxima ely hal o hei daily
o age om he na u al pas u e.
The e we e be ween 100 and 200 eindee a he
expe imen al si e and an addi ional 150 animals in he
su oundings. The e was no need o ence he si e
because he eindee s ayed he e o eed. The he de s
ed he animals daily wi h g ass silagehay and pelle s
on he clean snow o he plo s, anspo ing he o age by
snowmobile on a ack made o ha pu pose. The
eindee in he su oundings we e also ed. He de s
dis ibu ed abou one- hi d o he supplemen a y o age
(g ass silagehay and pelle s) o he plo s and sp ead
he emaining wo- hi ds in he a ea be ween he
plo s and in he su oundings o he expe imen al si e.
Acco dingly, all eindee *a maximum o 300 animals*
ecei ed supplemen al o age. On 15 June 2009 and on
26 May 2010, he le o e g ass silagehay, con aining
mos ly d y s aw, was emo ed using a ake om 10 o
he 20 plo s on which he he de s had ed eindee wi h
g ass silagehay).
Measu emen s
Vege a ion su eys. Vege a ion su eys we e con-
duc ed be o e he expe imen was es ablished (113
July 2008) and a e he i s (820 July 2009) and second
(412 July 2010) win e s. The su eys we e conduc ed
on wo subplo s (1 m
2
) pe plo (al oge he 250 plo s
100 subplo s) using a poin -coun ing me hod (Molau &
Table 1 The esul s o he analysis o g ass silage p oduced in summe 2008 and 2009 and pelle s used in he eindee eeding expe imen . The
analysis was ca ied ou by Vilja uuspal elu L d. (Mikkeli, Finland).
Pa ame e G ass silage 2008 G ass silage 2009 Ta
¨h i-Po o A ic pelle s Ta
¨h i-Po o 2 Balans pelle s
D y weigh (%) 20.4 24.8 87.0 87.2
C ude p o ein (%)
a
14.5 16.2 11.4 11.5
Raw ib e (%) 31.2 25.7 11.3 10.8
Ca (p/kg)
a
5.5 4.6 13 12
P g/kg
a
3.1 2.8 4.0 4.4
Mg g/kg
a
2.7 3.0 3.9 3.8
K g/kg
a
18 29 8.8 8.6
Na g/kg 0.21 0.21 1.8 2.6
Cu mg/kg
a
8.1 5.3 21 19
Mn mg/kg 76 86
Zn mg/kg
a
33 34 110 120
Fe mg/kg 88 63 170 190
To al con en o bac e ia c u/g 530 000 2100 B1000
Ca/P 1.8 1.7 0.2 0.2
Yeas con en c u/g 77000 B100 B100
pH 3.9
Suga % 8.4
a
Analysis in acco dance wi h Finnish Acc edi a ion Se ice s anda d ISO/IEC 17025.
Impac s o win e eeding o eindee on ege a ion and soil M. Tu unen e al.
4
(page numbe no o ci a ion pu pose) Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610
Molgaa d 1996). In addi ion o de e mining he co e age
o ascula plan s, lichens, mosses and hepa ics, he
su ey eco ded co e age o eindee aeces (see, e.g.,
Campbell e al. 2004; Ska in 2007; Moen e al. 2009),
unea en o age and ba e soil. In addi ion, he heigh s o
dwa sh ubs and Deschampsia lexuosa we e measu ed. To
a oid pseudo eplica ion, mean alues o each plo we e
calcula ed on he basis o he espec i e subplo s.
Plan and soil analysis. Since he chemical composi-
ion o plan s was expec ed o e lec eu ophica ion
caused by expe imen al eindee eeding be o e any
changes in soil and species co e age, he e e g een dwa
sh ub E. nig um L. (cu en -yea shoo s) and he decid-
uous V. my illus L. (cu en lea es) we e chosen as
indica o species. Samples we e collec ed om he 50
plo s on 25 July 2008, 21 July 2009 and 23 July 2010 o
he measu emen o C, H and N con en (LECO-1000,
Leco Co po a ion, S . Joseph, MI, USA). The measu e-
men s we e conduc ed om d y ash samples using a
he mal g a ime ic analyse wi h induc i ely coupled
plasma emission spec ome y (TJA I is Ad an age,
The mo Ja ell Ash Co po a ion, F anklin, MA, USA).
Soil samples o he measu emen o pH and C, H and
N con en we e aken om below he humus laye on
each expe imen al plo ; pH was measu ed in he ield
(Model Tes o 206, Lenzki ch, Ge many). Soil samples
we e d ied a 408C, p e- ea ed wi h a 2 mm sie e, and
analysed o C, H and N con en (% d y weigh ) using he
LECO-1000 ins umen .
S a is ical analysis
We used an analysis o co a iance o analyse he e ec o
eindee win e eeding on he N concen a ion and C/N
o E. nig um and V. my illus, he co e age (%) o species
in he g ound and ield laye , as well as he pH, N con en
and C/N o he soil. The da a om 2009 (a e he i s
win e ) and 2010 (a e he second win e ) we e ana-
lysed sepa a ely, because we we e in e es ed in inding
ou when he o e all impac s o win e eeding on
ege a ion and soil i s become isible. The explana o y
a iables used in he yea ly model we e eeding p ac ice
( i e le els) and he s a ing alues om 2008 (be o e he
expe imen was es ablished), he la e being used as
a co a ia e. In addi ion, we s udied he ela ionship
be ween he co e age o aeces and he co e age and
heigh o D. lexuosa by adding he co e age o aeces
in o he model and examining only g azed plo s ( ou
eeding p ac ice le els). The e was no need o make
ans o ma ions o he esponse a iables, because he
diagnos ic esidual plo s we e checked and ound o
be sa is ac o y.
The es s o he e ec s o eeding p ac ice and yea
2008 (co a ia e) a e p esen ed in ables o 2010 only.
The mul iple compa isons o he co a ia e-adjus ed
eeding p ac ice means we e analysed using he Tukey
K ame me hod (o e all signi icance le el: P0.05). The
eeding p ac ice means in 2008 and he co a ia e-
adjus ed means in 2009 and 2010 (wi h he mul iple
compa isons) a e p esen ed in he igu es o he species
whose co e age in 2010 exceeded 5%. The MIXED
p ocedu e o he SAS s a is ical so wa e package ( e sion
9.2) was used in he modelling analyses.
Resul s
Reindee ac i i y and le o e o age
Mean co e age o aeces, which indica es eindee
ac i i y on he plo s, was highe a e he i s win e
(2009) han a e he second (2010; Fig. 2a). A e bo h
yea s, he mean co e age o aeces was highes on he
plo s whe e eindee we e ed wi h pelle s (9% in 2009
and 5% in 2010), anging be ween 2 and 5% o o he
eeding p ac ices. The e we e no aeces on he enced
con ol plo s.
The unea en g ass silagehay on he plo s mainly
con ained s aw. The mean co e age o le o e s on
unclea ed plo s was 26% a e he i s yea (2009) and
29% a e he second yea (2010) as agains 5% and 7%
on he clea ed plo s, espec i ely (Fig. 2b). The e we e no
le o e s on con ol plo s.
Vege a ion su eys
In 2008, be o e any eeding p ac ices we e in oduced,
he ield laye o he pine-domina ed sub-xe ic hea h
o es si e was cha ac e ized by e e g een dwa sh ubs
(61%), deciduous dwa sh ubs (14%) and g aminoids
(14%). The mos common e e g een sh ub species we e
E. nig um (39%), Calluna ulga is (16%) and Vaccinium
i is-idaea L. (6%) and he mos common deciduous
sh ub was V. my illus L. (14%; Fig. 3ad). O he sh ubs,
among hem V. uliginosum L., Ledum palus e L., Diphasias-
um complana um (L.) Holub. and Lycopodium cla a um L.
we e a e (B1%). The only g aminoid was D. lexuosa
(14%; Fig. 3e). The mean pe cen age o he si e co e ed
by non- ege a ed ba e soil and li e was 9%. The g ound
laye exhibi ed mosses (73%), lichens (21%) and ba e
soil (6%; Fig. 3 i). The mos common moss and hepa ic
species we e Pleu ozium sch ebe i (34%), Dic anum spp.
(26%), Ba bilophozia spp. (7%), Hylocomium splendens
M. Tu unen e al. Impac s o win e eeding o eindee on ege a ion and soil
Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610 5
(page numbe no o ci a ion pu pose)

(5%) and Poly ichum spp. (B1%). O he lichens
p esen , he mos common species was Cladonia angi e a
(6%), a ype o win e o age p e e ed by eindee .
No in asi e plan species in oduced h ough g ass
silagehay could be obse ed on he plo s a e wo
win e s o simula ed eeding. The o e all e ec o
eeding, including he e ec o o age as well as eindee
on he pas u e, did no esul in signi ican di e ences in
he mean co e age o dwa sh ub species a e he i s
win e (2009; Fig. 3ad). A e he second win e (2010),
howe e , he e we e signi ican di e ences be ween
he eeding p ac ices in he mean co e age o he
dwa sh ubs E. nig um,C. ulga is and V. i is-idaea. The
co e age was lowe o ended o be lowe on plo s whe e
eindee we e ed wi h g ass silagehay (le o e s
no clea ed) compa ed o he con ol plo s (Fig. 3a, b,
d, Table 2). In con as , he e was a signi ican e ec o
eeding p ac ice on he co e age o he g aminoid D.
lexuosa a e he i s win e , in 2009 ( eeding p ac ice:
F3.09, P0.025; Fig. 3e): i s co e age inc eased on
plo s whe e eindee we e ed wi h g ass silagehay
(clea ed) when compa ed o enced con ols. A e he
second win e (2010), he heigh o D. lexuosa inc eased
signi ican ly on hese plo s ega dless o whe he le -
o e s we e clea ed when compa ed o enced o no
enced con ols ( eeding p ac ice: F8.33, PB0.0001;
da a no shown in he igu es). The models, in which
bo h he s a ing alues om 2008 and he eeding
p ac ices we e aken in o conside a ion, indica ed ha
he e was a signi ican posi i e linea ela ionship on
he ded plo s ( enced con ol plo s excluded) be ween he
co e age o aeces and he co e age and heigh o D.
lexuosa. In 2009, he eg ession coe icien o he
model o co e age o D. lexuosa was 0.32 (P0.030)
and o heigh 0.55 (P0.032). In 2010, he co espon-
ding eg ession coe icien s we e inc eased o 0.59
(P0.021) and 0.96 (P0.039), espec i ely.
The e was a signi ican e ec o eeding p ac ice on he
co e age o low-g owing mosses and hepa ics om he
i s win e , in 2009 (F10.96, PB0.0001) onwa ds (Fig.
3 h, Table 2), because le o e o age co e ed hem o a
conside able ex en . Fo example, he mean co e age o
Dic anum spp. a e he second win e (2010) was 28% on
clea ed g ass silagehay plo s, bu only 15% on un-
clea ed ones (Fig. 3h, Table 2). The e we e no signi ican
di e ences in co e age o lichens be ween he eeding
p ac ices a e he i s yea (2009). A e he second yea
(2010), howe e , he unclea ed plo s whe e eindee
we e ed wi h g ass silagehay o pelle s had a sig-
ni ican ly lowe co e age o lichens when compa ed o
un enced con ol plo s (Fig. 3i, Table 2).
Plan and soil analysis
A e he i s win e (2009), he e was a signi ican
e ec in N con en o E. nig um due o eeding p ac ices
(F4.68, P0.003). The mean N con en o E. nig um
was 79% (Fig. 4a), and he C/N a io was 89% highe
on he plo s wi h g ass silagehay eeding (le o e s no
clea ed) compa ed o con ol plo s ( enced/un enced).
A e he second win e (2010), he di e ences due o
eeding p ac ices we e mo e p onounced: he mean N
con en o E. nig um was 1114% highe on unclea ed
plo s and 913% highe on clea ed plo s s. con ol plo s
(un enced/ enced). Feeding eindee wi h pelle s did no
Fig. 2 The mean (wi h s anda d e o ) co e age (%) o (a) aeces and (b)
unea en g ass silagehay in he eeding expe imen a e he i s
win e (2009) and he second (2010), n10 plo s pe eeding p ac ice.
Impac s o win e eeding o eindee on ege a ion and soil M. Tu unen e al.
6
(page numbe no o ci a ion pu pose) Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610
inc ease he N con en o E. nig um lea es du ing he
expe imen (Fig. 4a, Table 3).
The e we e no signi ican di e ences among he
eeding p ac ices in he N con en o V. my illus a e
he i s win e (2009), bu a e he second win e
(2010), he N con en o V. my illus was 1315% highe
on unclea ed g ass silagehay plo s and 1416% highe
on pelle plo s when compa ed o con ols (un enced/
enced); Fig. 4b, Table 3).
No signi ican e ec s o eeding p ac ice on pH o on C
and N con en measu ed om he soil below he humus
laye could be obse ed du ing he expe imen . The only
excep ion was soil C/N, which exhibi ed signi ican
e ec s due o eeding p ac ices a e he second win e :
he a io was 27% lowe on g ass silagehay plo s
(le o e s clea ed) when compa ed o enced con ol
plo s (Fig. 4c, Table 3).
Discussion
No in asi e species in oduced om he g ass silage o
hay could be obse ed on he plo s du ing ou 2-yea
expe imen . This mos p obably shows ha he g ass
silagehay, whose main componen s a e Fes uca p a ensis
Deschampsia cespi osa,Phleum p a ense and Ta axacum
o icinale, was cu be o e seeding, ha he me hod o
p oducing he silage dec eased seed su i al o he seeds
ailed o ge mina e o some o he eason, o ha he
seedlings had los hei iabili y due o win e os s o
unsui able soil. Ano he eason o he lack o in asi e
Fig. 3 The mean (wi h s anda d e o ) co e age (%) o (a) Empe um nig um, (b) Calluna ulga is, (c) Vaccinium my illus, (d) Vaccinium i is-idaea and
(e) Deschampsia lexuosa in he ield laye , and ( ) all mosses and hepa ics, (g) Pleu ozium sch ebe i, (h) Dic anum sp. and (i) all lichens in he g ound
laye o a sub-xe ic hea h o es be o e he eeding expe imen was s a ed (2008), a e he i s win e (2009) and he second (2010). The igu es o
2008 a e s a ing means, and hose o 2009 and 2010 co a ia e-adjus ed means (2008 alues as a co a ia e) o eeding p ac ices. The di e en le e s
deno e di e ences be ween he eeding p ac ices a signi icance le el P0.05, n10 plo s pe eeding p ac ice.
M. Tu unen e al. Impac s o win e eeding o eindee on ege a ion and soil
Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610 7
(page numbe no o ci a ion pu pose)
species migh be he a he mode a e le el o supple-
men a y g ass silage/hay eeding (ca. 150 g pe m
2
pe
day) in combina ion wi h he sho du a ion o he
expe imen (110 days in o al du ing wo win e s in
200910). Ou ea lie esea ch on h ee in ensi ely used
eindee win e eeding pens (1360 eindee in 0.1100
ha pens o 410 win e s) in no he n Finland indica es
ha lichens, mosses and dwa sh ubs become eplaced
by as e -g owing g aminoids and he bs; hese include N-
a ou ing pionee species such as Epilobium angus i olium
and weeds such as S ella ia media ha a e in oduced
ei he h ough eindee o age (g ass silagehay) o
om he su oundings o he win e eeding pens
(Tu unen e al. 2010; Tu unen & Vuojala-Magga 2011).
The lack o in asi e species in he p esen s udy may also
be due o he esis ance o he sub-xe ic hea h o es o
he o e all impac s o o age and eindee ; he appea -
ance o in asi e species is dependen on he esis ance o
he habi a ype and/o i s abili y o esis o eco e om
e osion caused by g azing, ampling and digging. The
habi a ypes ha ha e he weakes esis ance include
mois mi es and xe ic hea h o es s, whe eas mesic and
sub-xe ic hea h o es s* he si e ypes in he p esen
s udy*a e mo e esis an (Kelloma
¨ki & Saas amoinen
1975; Ukkola 1995).
Ou expe imen showed a signi ican o e all impac o
win e eeding on he ield and g ound-laye ege a ion
o a sub-xe ic pine-domina ed hea h o es in no he n
Finland. Since N is he limi ing ac o o plan g ow h in
nu ien -poo hea h o es s (Vi ousek & Howa h 1991),
bo h he as -g owing, N-demanding g aminoid D. lex-
uosa and he slow-g owing e e g een dwa sh ub E.
nig um and deciduous dwa sh ub V. my illus we e able
o u ilize he inc eased a ailabili y o N compounds. In D.
lexuosa, his could be obse ed al eady a e he i s
win e bo h as co e age and heigh inc easing wi h he
co e age o eindee aeces and as inc eased co e age
whe e g ass silagehay had been used. In E. nig um,N
con en inc eased on g ass silagehay plo s (le o e no
clea ed in he sp ing) a e he i s win e , and in V.
my illus a e he second win e . Ea lie s udies ha e
indica ed ha he impac s o eindee g azing on ege a-
ion and soil depend g ea ly on eindee popula ion
densi y, season o g azing and he ype and successional
s a e o he si e (Va
¨ e e al. 1995; Va
¨ e e al. 1996;
Suominen & Olo sson 2000; Moen e al. 2009; S a k e al.
2010). In oligo ophic climax-s age Cladonia s ella is- ich
pine-domina ed o es s in no he n Finland, hea y g az-
ing has been ound o esul in de elopmen o he
unde s o ey ege a ion in wo di ec ions: owa ds i s
Cladonia s age, o owa ds b yophy e- ich si es ha ing
meso ophic ea u es such as ele a ed soil N con en and
pH (Va
¨ e e al. 1995; Va
¨ e e al. 1996). The s udies o
hea hlands and und a in no he n Fennoscandia ha e
shown ha eindee may inc ease he abundance o
g aminoids and p oduc i i y o ege a ion h ough se -
e al mechanisms: a ou ing plan s wi h high-quali y
li e (Olo sson & Oksanen 2002), p oducing apidly
ecycled u ine and aeces ( an de Wal e al. 2004) and
inc easing soil empe a u es (Olo sson e al. 2004). In
gene al, mode a e g azing has been ound o inc ease
di e si y on hea hlands, whe eas di e si y has emained
unal e ed o sligh ly lowe on mode a ely g azed g ass-
and he b-domina ed meadows (Suominen & Olo sson
2000).
Ou esul s show ha bo h eindee d oppings ( aeces
and u ine) and unea en o age le on he pas u e
con ibu ed o he inc eased a ailabili y o N compounds
in he sub-xe ic o es s udied. Reindee aeces and u ine
p o ide highly decomposable esou ces ha a e ich in
labile nu ien s, such as ammonium and ni a e, and can
s imula e soil mic obial biomass and ac i i y, ne C and N
Table 2 F- es s o he e ec s o eeding p ac ice (10 plo s pe eeding p ac ice) and he yea 2008 (da a measu ed be o e he expe imen was es a-
blished) on he species co e age in he g ound and ield laye o he sub-xe ic hea h ype o es a e wo win e s o expe imen al eeding o eindee in
2010.
Feeding p ac ice 2008 (co a ia e)
FSigni icance FSigni icance
Field laye
Empe um nig um 5.26 0.002 199.90 B0.0001
Calluna ulga is 5.63 0.001 1510.28 B0.0001
Vaccinium my illus 1.20 0.323 108.96 B0.0001
V. i is-idaea 2.36 0.068 4.63 0.037
Deschampsia lexuosa 8.70 B0.0001 142.24 B0.0001
G ound laye
Mosses and hepa ics 9.03 B0.0001 20.57 B0.0001
Pleu ozium sch ebe i 3.45 0.016 211.84 B0.0001
Dic anum sp. 5.44 0.001 123.15 B0.0001
Lichens 3.75 0.010 39.02 B0.0001
Impac s o win e eeding o eindee on ege a ion and soil M. Tu unen e al.
8
(page numbe no o ci a ion pu pose) Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610
mine aliza ion, as well as plan N acquisi ion and g ow h
(Suominen & Olo sson 2000; S a k e al. 2003; an de
Wal e al. 2004). The o igin o he N compounds de i ed
om he unea en le o e o age may be na u al, ha
is, a species o o age plan , o a i icial, o example,
e ilize s, p ese a i es and/o addi i es used du ing
he p oduc ion o g ass silage. P o eolysis (b eakdown
o p o eins) in he g ass silage causes ammonium-N o be
o med, he con en o which inc eases pa icula ly when
he suga con en o g ass silage is low (McDonald e al.
1991; Aagnes & Ma hiesen 1995; Heiska i & Nieminen
2004). The N compounds, pa icula ly ammonium-N,
may hus ha e been abso bed by he ege a ion and soil
(o adso bed on o hem) unde he o age le o e ,
leading o inc eased issue N con en in E. nig um and
V. my illus lea es. The elease o nu ien s om eindee
aeces and unea en o age le on he pas u e is g ea ly
dependen on soil mois u e (see, e.g., an de Wal e al.
2004), mos p obably being g ea es immedia ely a e
snowmel (17 May in he I alo*Ku u a a ea in an
a e age yea (da a ob ained om he Finnish Me eo -
ological Ins i u e) and a e pe sis en ains.
O he eeding p ac ices applied, eeding wi h g ass
silagehay (le o e no clea ed in he sp ing) mos
o en caused he ea lies and mos signi ican o e all
impac s on ege a ion. The low-g owing lichens and
mosses, such as Dic anum spp., and dwa sh ubs ob-
se ed in his s udy a e easily co e ed by eindee o age
le o e s on pas u e. Besides being an addi ional N
sou ce, unea en o age le in hick laye s on pas u e
may di ec ly dec ease pho osyn hesis and hus supp ess
he g ow h o g ound- and ield-laye ege a ion. In he
sp ing, a hick laye o unea en o age wi h a high s aw
con en may insula e he soil om hea and lead o la e
mel ing o soil os and lowe soil empe a u es (see Pe h
& Ho n 2006). Con e sely, o age le o e s may e en ou
he empe a u e a ia ion o he soil and slow e apo a-
ion om he unde lying soil and he ege a ion, he eby
inhibi ing he exchange o o he gases also.
The p esen s udy shows ha he o e all impac o
eeding eindee wi h pelle s on ege a ion and soil
below he humus laye was less signi ican han ha
caused by eeding wi h g ass silagehay. On he plo s
whe e eindee we e ed wi h pelle s, he species co e age
o dwa sh ubs and mosses and hepa ics emained
unchanged, and N con en s in E. nig um lea es we e
lowe han on he plo s whe e eindee we e ed wi h
g ass silagehay. Reindee p e e pelle s o hay o g ass
silage, and le o e s a e no common. The e o e, he
p ocessing o c ude hay*in o hay pelle s, o example*
and/o he use o comme cial g ain-based eindee
pelle s could minimize he amoun o le o e o age
Table 3 F- es s o he e ec s o eeding p ac ice (10 plo s pe eeding p ac ice) and yea 2008 (da a measu ed be o e he expe imen was es ablished) on N concen a ion (%) and C/N in E. nig um,
V. my illus and soil, and soil pH a e wo win e s o expe imen al eeding o eindee in 2010, as well as he mul iple compa isons (TukeyK ame me hod) o he co a ia e-adjus ed eeding p ac ice
means.
Feeding p ac ice 2008 (co a ia e) Co a ia e-adjus ed eeding p ac ice means (s anda d e o ) 2010
FSigni icance FSigni icance Con ol ( enced) Con ol
G ass silagehay, no
clea ed
G ass silagehay,
clea ed Pelle s
E. nig um
N 9.94 B0.0001 0.29 0.591 1.26 (0.03)
a
1.31 (0.03)
a
1.47 (0.03)
b
1.44 (0.03)
b
1.36 (0.03)
ab
C/N 10.84 B0.0001 0.45 0.506 43.99 (0.85)
a
42.42 (0.85)
a
37.28 (0.84)
bc
38.4 (0.84)
b
40.76 (0.84)
ab
V. my illus
N 6.14 0.0005 4.96 0.031 1.46 (0.05)
a
1.49 (0.04)
a
1.68 (0.04)
b
1.57 (0.04)
ab
1.70 (0.04)
b
C/N 6.90 0.0002 7.80 0.008 36.16 (0.95)
a
35.00 (0.83)
a
31.34 (0.83)
b
33.40 (0.87)
ab
30.79 (0.84)
b
Soil
N 0.38 0.818 0.94 0.339 0.068 (0.01) 0.064 (0.01) 0.075 (0.01) 0.066 (0.01) 0.069 (0.01)
C/N 2.69 0.044 0.07 0.793 23.87 (1.50)
a
19.34 (1.50)
ab
21.90 (1.50)
ab
17.40 (1.54)
b
19.68 (1.60)
ab
Ph 1.03 0.403 3.18 0.082 4.61 (0.10) 4.75 (0.10) 4.66 (0.10) 4.87 (0.10) 4.68 (0.10)
a,b,c
Di e ences be ween he eeding p ac ices a signi icance le el P0.05.
M. Tu unen e al. Impac s o win e eeding o eindee on ege a ion and soil
Ci a ion: Pola Resea ch 2013, 32, 18610, h p://dx.doi.o g/10.3402/pola . 32i0.18610 9
(page numbe no o ci a ion pu pose)