Vol.:(0123456789)
1 3
Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
h ps://doi.o g/10.1007/s10342-019-01163-9
ORIGINAL PAPER
Shi ing ome en‑aged managemen olessin ensi e o es y
in a ying p opo ions o  o es land inFinland: impac s onca bon
s o age, ha es emo als, andha es ing cos s
Ja iVauhkonen1 · TuulaPackalen1
Recei ed: 4 June 2018 / Re ised: 24 Decembe 2018 / Accep ed: 8 Janua y 2019 / Published online: 21 Janua y 2019
© The Au ho (s) 2019
Abs ac
Many s udies ha e epo ed inc eased mul i- unc ionali y and inancial p o i s due o a shi om e en- o une en-aged o es
managemen . Howe e , li le is known ( om long- e m expe iences o p edic ions) how al e na i e managemen sys ems
could a ec na ional-scale wood p oduc ion and ca bon s o age, i adop ed o e e y la ge a eas. We analysed hese e ec s
using an a ea-based amewo k, in which mul iple Ma ko chain models we e used o simula e he de elopmen o o es s
acco ding o di e en managemen sys ems. Classi ica ion o o es s o wood a ailabili y ca ego ies was used o de e mine he
sys em o be applied. We enhanced he amewo k o allow shi s be ween managemen sys ems ha co espond o en o ced
o olun a y changes in o es use. Simula ions o ex ensi e shi s om con en ional e en-aged managemen o al e na i e
sil icul u al sys ems e ealed in e es ing de elopmen al pa e ns ha canno be di ec ly deduced om s udies ha upscale
om smalle a eas. Ou esul s show ha he amoun o ca bon s o ed by Finnish o es s can be inc eased by applying less
in ensi e managemen sys ems, al hough his has ade-o s in e ms o ha es s and associa ed inancial cos s. The le el o
ade-o s di e ed depending on he ype o o es ha shi ed be ween managemen sys ems and whe he a eas we e also
assumed o be comple ely se aside om o es y. These di e ences we e u he p onounced i he desi ed ha es le els
and hei alloca ion changed along wi h he managemen sys em. I he s udied a ibu es we e conside ed a he same ela-
i e scale and wi h equal weigh ing, he ex ensi e shi s o al e na i e managemen sys ems exhibi ed he s onges impac
on ha es ing cos s.
Keywo ds Eu opean Fo es y Dynamics Model (EFDM)· Fo es esou ce p ojec ion· In eg a ed o es managemen ·
Ma ix model· Na ional Fo es In en o y (NFI) da a· Scena io analysis
In oduc ion
Cu en ly, se e al in e na ional and na ional s a egies
s i e o a ansi ion om a ossil- o a bio-based economy,
which calls o an inc ease in he use o ( o es ) biomass o
p oduc s, such as bio uels and ene gy, chemicals, polyme s,
and wood-based s uc u es. Howe e , conce ns in ega d o
agmen a ion, deg ada ion, and loss o o es habi a s ha e
been inc easingly exp essed and in e na ional and na ional
ag eemen s ha e been signed o e e se his end. In pa -
icula , he es o a ion o a leas 15% o deg aded ecosys-
ems by 2020 (CBD 2010), also known as Aichi Ta ge 15,
is widely accep ed as a na ional and global conse a ion
a ge . Howe e , main aining bo h high economic o es
yields and he iabili y o o es species in ol es ade-o s.
In Finland, o example, Ko iaho e al. (2016) ha e es ima ed
ha mee ing Aichi Ta ge 15 would cos be ween 12 and 23
billion eu o, o 368–658 million eu o pe annum, when only
o es s and pea lands a e conside ed and i he conse a ion
ac ions we e p olonged un il 2050. Fo compa a i e pu -
poses, annual s umpage ea nings amoun o app oxima ely
1.5 billion eu o (MAF 2015).
Communica ed by Mi en del Rio.
Elec onic supplemen a y ma e ial The online e sion o his
a icle (h ps ://doi.o g/10.1007/s1034 2-019-01163 -9) con ains
supplemen a y ma e ial, which is a ailable o au ho ized use s.
* Ja i Vauhkonen
ja i. auhk[email p o ec ed]
1 Bioeconomy andEn i onmen Uni , Na u al Resou ces
Ins i u e Finland (Luke), Yliopis oka u 6, 80100Joensuu,
Finland
220 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
Ra he han seg ega ing he o es and conse a ion sec-
o s, a good comp omise could be achie ed by combining
cos -e ec i e conse a ion ac ions wi h he sus ainable use
o o es s and pea lands (see also Ko iaho e al. 2016). Fo
example, he managemen o a p opo ion o o es land
as “mul i-use conse a ion landscapes” (MUCLs; Hanski
2011) is a p ac ical and cos -e ec i e means o aid o es
conse a ion s a egies, in addi ion o he s ic p o ec ion o
o es ne wo ks (see also S e ens and Mon gome y 2002).
Mo e p ecisely, Hanski (2011) has p oposed ha manage-
men o a hi d o he land as MUCLs, and he s ic p o-
ec ion o a hi d o his a ea, would be su icien o he
conse a ion needs o specialis species, pa icula ly i he
MUCLs we e composed o agg ega ed habi a pa ch clus e s
and could be connec ed o exis ing p o ec ed a ea ne wo ks
(see also Rybicki and Hanski 2013). Howe e , he wo k ci ed
abo e does no speci y easible sil icul u al p ac ices o
managemen in ensi ies o he p opo ion o MUCLs ( wo-
hi ds) ha a e no s ic ly p o ec ed. Ye , we can assume
ha such a eas could be managed acco ding o he p inciples
o mul iple-use (Fü s enau e al. 2007) o in eg a ed o es
managemen (Diaci e al. 2011), which suppo he p oduc-
ion o ecosys em se ices o he han jus species conse a-
ion based on a mo e di e se se o sil icul u al p ac ices,
compa ed o con en ional o a ion o es y o e en-aged
managemen sys ems.
Fo es y p ac ices in Finland ha e been based on e en-
aged managemen since Wo ld Wa II (Kuulu ainen e al.
2012), bu op ions o o es managemen p ac ices will
clea ly inc ease in he u u e. By op ions, we e e o he a -
ious o ms o une en-aged managemen , such as con inuous
co e o es y as de ined by Pukkala (2016a). Con inuous
co e o es y essen ially di e s om e en-aged manage-
men in ha i a oids clea elling and plan ing by u ilizing
hinnings om abo e and by p omo ing na u al egene a ion.
These choices may con e s ands owa ds une en-aged o -
es s uc u es, al hough con e ging o a s eady-s a e s uc u e
o any kind is no equi ed (Pukkala 2016a). Con inuous
co e o es y is expec ed o become mo e common, because
o i s po en ial o supply mul iple ecosys em se ices (Puk-
kala 2016b; Peu a e al. 2018) and educe he inancial cos s
ela ed o egene a ion and o he sil icul u al ope a ions
(Pukkala 2016a) compa ed o e en-aged managemen (see
also Knoke 2012; Kuulu ainen e al. 2012; Pue mann e al.
2015; Nieminen e al. 2018).
Compa isons o al e na i e o es managemen sys ems
and subsequen ade-o analyses a e ypically based on
long- e m obse a ions (Su he land e al. 2016; S eng-
bom e al. 2018), me amodelling (La ond e al. 2017) o
simula ions. The la e ha e been ca ied ou a he o es
s and o small o es holding le el (Pukkala e al. 2011;
Pukkala 2016a, b; Ca pen ie e al. 2017), he landscape
le el (> 100km2; T i iño e al. 2015; Diaz-Bal ei o e al.
2017; Peu a e al. 2018), and he egional le el (> 1000km2;
Sch ö e e al. 2014; Pang e al. 2017). Howe e , long- e m
expe iences o p edic ions as o how al e na i e p ac ices
may a ec na ional-scale wood p oduc ion i adop ed o e
e y la ge a eas a e no known. Wi h he excep ion o
ecen ly o mula ed g ow h (Bollandsås e al. 2008; Puk-
kala e al. 2013) and hinning models (Pukkala e al. 2015;
Vauhkonen and Pukkala 2016), mos con en ional o es
simula o s and p ojec ion ools ha e been de eloped o
e en-aged o es y sys ems. Con en ional models o o es
de elopmen would, he e o e, ex apola e ou side he o igi-
nal popula ion i applied in une en-aged o es s. Mo eo e ,
de ailed o es -speci ic hinning ins uc ions migh no ec-
oncile wi h la ge-a ea p ojec ions based on agg ega ed spa-
ial scales (c ., Ve ke k e al. 2014; C eu zbu g e al. 2017;
Mouche e al. 2017). As such, he e is a need o lexible
ools ha can combine de ailed ins uc ions wi h p ojec ion
capabili ies o la ge a eas.
F om he pe spec i e o egional o na ional-le el wood
p oduc ion, he a eas subjec o conse a ion o in eg a ed
managemen educe he amoun o Fo es s A ailable o
Wood Supply (FAWS; Albe di e al. 2016). F om he poin
o iew o p o isioning o non-wood o es p oduc s o o he
ecosys em se ices, i is use ul o also simula e he de elop-
men o a ea and g owing s ock in he emaining a eas—i.e.
in Fo es s No A ailable o Wood Supply (FNAWS) and
Fo es s wi h Res ic ions on A ailabili y o Wood Supply
(FRAWS; see also Vauhkonen and Packalen 2017). Ex en-
si e o es in en o ies, such as he Na ional Fo es In en o y
(NFI), also p o ide da a o he simula ion o ansi ions
due o g ow h o managemen o o es s p io i ized o
uses o he han solely wood p oduc ion. Se e al app oaches
ha e been p esen ed o he p ojec ion o he u u e de el-
opmen o o es esou ces based on ansi ion p obabili y
ma ices o s and-speci ic diame e classes (e.g. Bollandsås
e al. 2008; Schou and Meilby 2013; Roessige e al. 2016).
Howe e , co esponding Ma ko chain models based on
ansi ion ma ices o o es size and s uc u e classes (e.g.
Vauhkonen and Packalen 2017) could be mo e sui able o
a ea-based p ojec ions o o es dynamics based on he NFI
da a.
Vauhkonen and Packalen (2017) simula ed he de elop-
men o o es size and s uc u e classes de i ed om NFI
da a by combining mul iple Ma ko chain models o di -
e en sil icul u al sys ems. In hei s udy, classi ica ion o
o es s o wood a ailabili y ca ego ies de e mined which
sys em was applied. Wood a ailabili y depended on admin-
is a i e o es use es ic ions as eco ded in he NFI da a.
A eas whe e o es y ope a ions we e o bidden (FNAWS)
o es ic ed o selec i e ha es s o he enhancemen o
ecosys em se ices o he han wood supply (FRAWS) we e
dis inguished om FAWS. The u u e de elopmen o FAWS
was simula ed wi h e en-aged managemen ; FRAWS wi h
221Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
con inuous co e o es y, in which he inal elling was
eplaced by hinning om abo e; and only he na u al p o-
cesses we e simula ed o FNAWS. The assump ion ha
es ic ions on o es use de e mine he sil icul u al sys em is
a simpli ica ion ha migh be un ealis ic, especially, i la ge
p opo ions o FAWS di e ge om e en-aged managemen
in he u u e, as easoned abo e. Howe e , he FAWS in ou
in en o y da a a e la gely composed o e en-aged s ands;
o es managemen ins uc ionswe e based on e en-aged
sil icul u e un il he inal measu emen s o he 11 h Finnish
NFI in 2013. Thus, inc easing he p opo ion o al e na i e
managemen sys ems in FAWS undamen ally means ha
mo e o es s will be managed in a simila ashion o FRAWS
o FNAWS a e obliga o y managed because o adminis a-
i e o es use es ic ions. We p opose ha simula ion o
shi s be ween hese ca ego ies should allow u he exami-
na ion o he na ional-scale e ec s o mo ing om e en-
aged, o a ion o es managemen o al e na i e sil icul u al
egimes, which may esul om en o ced poli ical decisions
o p omo e less in ensi e o es y o inc eased na u e conse -
a ion o olun a y changes in he use o o es s.
The objec i es o his s udy a e (1) o enhance he
Ma ko chain modelling amewo k by he simula ion o
shi s be ween o es managemen sys ems and (2) o use
he de eloped amewo k o assess he ade-o s be ween
ca bon s o age, ha es emo al, and ha es ing cos s due
o hese shi s. We i s de i ed benchma k p ojec ions o
he u u e de elopmen o o es s acco ding o sil icul u al
sys ems associa ed wi h cu en wood a ailabili y ca ego ies
as desc ibed in he p e ious pa ag aph. We hen epea ed
he simula ions wi h a ying p opo ions o e en-aged o -
es s managed acco ding o con inuous co e o es y o se
aside o assess he e ec s o di e ging om business-as-
usual managemen in an inc easing p opo ion o o es land.
Me hods
Simula ion amewo k andda a
We simula ed he u u e de elopmen o o es s in Finland
using an a ea-based Ma ko chain model (c . Vauhkonen
and Packalen 2017). Ou analyses conside ed almos he
en i e o es land o Finland, excluding no he nmos a eas
and o es s loca ed in he sou he n a chipelago, which a e
conside ed o ha e a low impo ance in e ms o wood sup-
ply o Finland. The o al a ea o o es s on p oduc i e and
poo ly p oduc i e o es land was 21.28 million ha, wi h an
ini ial g owing s ock o 2234 million m3 ( he 11 h Finnish
NFI–NFI11; measu emen s in 2009–2013), which equa es
o app oxima ely 95% o he en i e g owing s ock in Finland.
As in Vauhkonen and Packalen (2017), he o es s we e
assigned o wood a ailabili y ca ego ies acco ding o o es
use es ic ions in he NFI da a: ini ially, 10.1% and 10.6% o
he o al o es a ea we e classi ied as FNAWS and FRAWS,
espec i ely, wi h he emainde classi ied as FAWS. In o de
o s udy he na ional-scale e ec s o di e ging om he
cu en managemen sys em, a ying p opo ions o o es
we e emo ed om FAWS and assigned o ei he FRAWS
o FNAWS. These we e subsequen ly simula ed o u u e
de elopmen along wi h o es s ha a e cu en ly assigned
o hese ca ego ies due o adminis a i e o es use es ic-
ions. Speci ically, o es s ha ecei ed e en-aged manage-
men in he benchma k simula ions we e ei he assigned o
con inuous co e o es y (in he p opo ion ansi ed o he
FRAWS ca ego y) o se aside (FNAWS) a he beginning
o he simula ions.
The selec ion o land ansi ed om FAWS o he o he
ca ego ies was simula ed acco ding o ou di e en s a e-
gies ha mimic he di e en d i e s and land a ailabili y
in he ansi ions. The aim was o mimic he es ablishmen
o ei he (a) MUCLs as a a ian o he concep p oposed
by Hanski (2011), which consis o bo h con inuous co e
o es y and se -aside a eas; o (b) mul i-use landscapes
(MULs) managed by con inuous co e o es y wi hou se -
aside a eas. The ou s a egies we e ob ained by emphasiz-
ing he selec ion o o es s wi h high (s a egies MUCLhigh,
MULhigh) o low (MUCLlow, MULlow) p oxy conse a-
ion alues, de e mined as a unc ion o he ma u i y o he
ees, ee species composi ion, and si e e ili y (Appendix
2; Leh omäki e al. 2015). Because o i s o mula ion, he
p oxy migh ac as a su oga e no only o he biodi e si y
ea u es o conse a ion in e es , bu also he mo e gene al
mul iple-use po en ial o a o es . As u he discussed in
Sec .4.1, he ou s a egies gene ally co e he ange o p o-
duc ion possibili ies, by educing he managemen in ensi y
in ex ensi e p opo ions o he di e en o es ypes.
Each land ansi ion s a egy was composed o eigh
p opo ions p, whe e p was ei he 5, 10, …, o 40% o he
FAWS land a ea, which was hen added o ei he FRAWS
o FNAWS. The p opo ions we e selec ed by i s compu -
ing he conse a ion alue (cons al) o each NFI plo , as
desc ibed in Appendix 2, and he cumula i e dis ibu ion
o cons al sepa a ely o each o he 15 o es y egions (o
Fo es Cen es as dis inguished by he Finnish NFI). The
numbe o plo s equi ed o ep esen p% o he FAWS land
a ea was selec ed a equal in e als om he cumula i e dis-
ibu ion o each o es y egion as ollows:
• MUCLhigh p% o FAWS was i s selec ed acco ding o
cons al. A hi d o his a ea was u he selec ed acco d-
ing o he cumula i e dis ibu ion o cons al compu ed
o plo s in he selec ed p% and assumed o ansi o
FNAWS (no managemen ). The emaining p opo ion o
p% was assumed o ansi o FRAWS and o be managed
as con inuous co e o es y. The emphasis in he new
222 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
MUCLs es ablished acco ding o his s a egy was on
FAWS wi h a high conse a ion alue. One- hi d o his
selec ion (wi h an emphasis on o es s wi h he highes
conse a ion alues) was se aside.
• MUCLlow As wi h MUCLhigh, bu he land a ea o
ansi was selec ed acco ding o he in e se o con-
s al, excluding plo s wi h cons al = 0 (i.e. ba e land).
The emphasis in he new MUCLs was on FAWS wi h a
low conse a ion alue. One- hi d o he o es s wi h he
highes conse a ion alues in he i s selec ion we e
comple ely se aside.
• MULhigh p% o FAWS was selec ed acco ding o con-
s al and assumed o ansi o FRAWS o be managed
as con inuous co e o es y. The emphasis in he new
MULs was on FAWS wi h high conse a ion alue. No
addi ional o es a eas we e se aside.
• MULlow As wi h MULhigh, bu he land a ea o ansi
was selec ed acco ding o he in e se o cons al, exclud-
ing plo s wi h cons al = 0 (i.e. ba e land). The emphasis
in he new MULs was on FAWS wi h low conse a ion
alue. No addi ional o es a eas we e se aside.
Simula ions o  o es de elopmen
As he Ma ko chain model, we used . 2.0. o he Eu opean
Fo es y Dynamics Model (EFDM), which is implemen ed
in he R s a is ical modelling en i onmen (R Co e Team
2016) and can be downloaded om h ps ://gi hu b.com/ec-
j c/e dm as open sou ce unde he Eu opean Union Public
License (EUPL). The EFDM app oach is based on a ang-
ing he o es a ea in o ma ix cells acco ding o ecological
and socioeconomic ac o s and simula ing he de elopmen
o he esul ing ma ices. In he simula ions, he a ea ep-
esen ed by each ma ix cell is managed acco ding o a se
o p e-de ined ac i i ies, which may ansi he a ea o o he
cell(s) depending on he ansi ion p obabili ies associa ed
wi h he ac i i ies. In p ac ice, he o es a ea dis ibu ion
a e one simula ion s ep (i + 1) is ob ained as a mul iplica-
ion o he a ea in s a e i by he p obabili y ha he a ea
ecei es one o j ac i i ies and he ac i i y-condi ional an-
si ion p obabili ies ( o de ails, see Si kiä 2012 o Packalen
e al. 2014).
The simula ions we e ca ied ou in i e-yea ime-s eps,
which co espond o he measu emen in e al in he da a
used o de i e he ansi ion p obabili ies. In o al, en s eps
we e simula ed (i.e. he las yea in he simula ion pe iod
is app oxima ely 2060, depending on he ini ial measu e-
men yea ). The ini ial o es s a e, (business-as-usual) an-
si ion and ac i i y p obabili ies, and ou pu coe icien s o
he model we e de i ed om he NFI11 da a acco ding o
he wo k low p esen ed by Vauhkonen and Packalen (2017)
based on using pe manen NFI plo s as pai wise da a o
na u al p ocesses. Appendix 2 also illus a es he p ocess
o de i ing he ini ial s a e and he simula ion o he u u e
de elopmen o one example o es . In he ollowing sec-
ions, we b ie ly desc ibe he pa ame e iza ion, bu no e ha
pa ame e s and hei e ec s a e explained in mo e de ail in
Vauhkonen and Packalen (2017).
Ini ial s a e
The EFDM is pa ame e ized by ma ices wi h dynamic
(e.g. age, olume) and s a ic (e.g. geog aphical egion,
si e e ili y) dimensions (hence o h “ ac o s”). Simila o
Vauhkonen and Packalen (2017), we de ined he dynamic
ac o s sepa a ely o o es s o be managed using di e en
sil icul u al a ibu es, such as age and olume o e en-
aged managemen sys ems o s em numbe and olume o
con inuous co e o es y. The na u al p ocesses o FNAWS
we e simula ed using age and olume ma ices; no di e -
ences would ha e been obse ed i s em numbe and ol-
ume ma ices we e applied (Vauhkonen and Packalen 2017).
The con inuous measu emen s om he NFI we e classi ied
using age classes o 0, 5, 10, …, 120, 120+ yea s, whe eas
he class limi s o bo h he olume and s em numbe we e
de e mined as he alues o he 10 h, 20 h, …, 90 h and 95 h
quan iles o he pai wise obse a ions made om he pe -
manen NFI plo s. The class limi s (p esen ed in Appendix
1) we e de ined by Vauhkonen and Packalen (2017) wi h a
mo i a ion o ob ain an app oxima ely equal amoun o pai -
wise obse a ions pe class and o educe he alue ange o
he las class by hal ing he numbe o obse a ions included
in i . These dynamic ac o ma ices we e de i ed sepa a ely
by applying he ollowing s a ic ac o s: (1) known land-use
es ic ions: FAWS, FRAWS, FNAWS; (2) o es owne ship:
p i a e, public + o he ; (3) si e e ili y: a o al o i e ca ego-
ies ha co espond o he ou axa ion classes ha p oduc-
i e o es s a e assigned in he Finnish NFI + a i h class
ha includes all poo ly p oduc i e o es land; (4) dominan
species: pine, sp uce, deciduous ees.
Managemen ac i i ies and hei p obabili ies
Possible managemen ac i i ies we e “no managemen ”
(i.e. simula ion o na u al p ocesses only), “ hinning”, and
“ egene a ion ha es ”. A “ hinning” always e e ed o a
managemen hinning and was implemen ed as a hinning
om below o bo h FAWS and FRAWS. A “ egene a-
ion ha es ” was implemen ed ei he as a inal elling in
he e en-aged managemen sys em (simula ed o FAWS)
o a hinning om abo e in con inuous co e o es y ( o
FRAWS), as desc ibed in Sec .2.2.3.
The p obabili ies o he ac i i ies we e de e mined in wo
s eps. Fi s , he NFI da a we e used o compu e he mu ual
p opo ions o he ac i i ies, esul ing in wo al e na i e allo-
ca ions o he ac i i ies: (1) a business-as-usual alloca ion
223Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
(ABAU) based on he p opo ions o pe manen NFI plo s no
managed, hinned, o egene a ed du ing he mos ecen
5-yea pe iod; o (2) a schoolbook-alloca ion (ASB) based
on he p opo ions o he NFI plo s ha should be managed
wi hin he nex 5yea s s ic ly acco ding o o es manage-
men ins uc ions. The p opo ions we e based on he NFI
eco ds ha we e made acco ding o he ins uc ions ha
p e ailed a he ime o he measu emen s [ o mo e de ails,
see Y jölä (2002)]. The wo al e na i e alloca ions a e class
speci ic (c . Figu e1 o Vauhkonen and Packalen 2018);
ollowing he ini ial p opo ions o ASB would, in gene al,
in ol e p oposing much mo e managemen ac i i ies han
Fig. 1 E ec s o wood a ail-
abili y on ca bon s o age a he
end o he simula ion pe iod
( op panel), o al ha es emo -
als du ing he simula ion pe iod
(middle), and a e age ha es -
ing cos s o e he simula ion
pe iod (bo om) unde business-
as-usual ac i i y and ansi ion
p obabili ies. The x-axes o he
sub- igu es indica e he a ea
di e ging om e en-aged man-
agemen . The black- illed do s
ep esen simula ions wi h he
cu en p opo ions o Fo es s
A ailable o Wood Supply
(FAWS), Fo es s No A ailable
o Wood Supply (FNAWS) and
Fo es s wi h Res ic ions on
A ailabili y o Wood Supply
(FRAWS). The lines depic
scena ios, whe e he p opo ion
o he FAWS a ea co espond-
ing o di e en le els o p%
was shi ed om FAWS o he
o he wo ca ego ies acco ding
o he ou s a egies desc ibed
in Sec .2.1. The open ci cles
deno e a hypo he ical si ua ion,
whe e all o es s a e conside ed
as FAWS and simula ed acco d-
ing o he e en-aged manage-
men sys em

224 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
ABAU. Second, he ac i i y p obabili ies we e i e a ed o he
le el ha yielded a speci ied o al oundwood ha es le el.
As in Vauhkonen and Packalen (2017), he i e a ion was
ca ied ou by epea edly mul iplying he ac i i y p obabili-
ies ≠ 0 by 1.01 o 0.99, depending on he sign o he di e -
ence be ween he goal and he ha es le el gi en by he cu -
en ac i i y p obabili ies, un il he ha es ing goal was me
o he ac i i y p obabili ies could no be changed. The speci-
ied o al ha es goal was calcula ed om wo ha es le els
p o ided in he Na ional Fo es S a egy o Finland (MAF
2015): he business-as-usual le el (65 million m3/a) and he
desi ed u u e le el (80 million m3/a). Because ou analyses
conside ed abou 95% o he o al g owing s ock in Finland,
bo h goals we e mul iplied by his p opo ion o yield he
inal ha es a ge s o app oxima ely 62 million m3/a o 76
million m3/a, espec i ely. Ac i i y p obabili ies we e i e -
a ed o mee hese ha es ing goals only a he beginning o
he simula ions. The ea e , he same p opo ion o he land
a ea was managed in e e y simula ion s ep, i.e. he olume
ha es ed a e he i s simula ion s ep depended on how
he o es class dis ibu ion e ol ed du ing he simula ions.
T ansi ion p obabili ies
The ansi ion p obabili ies o he na u al p ocesses (g ow h)
we e de i ed using pai wise obse a ions om he pe ma-
nen NFI plo s ( o al: 11,987 o abou 23% o he plo s),
which we e measu ed a app oxima ely i e-yea in e als
be ween NFI11 and he ea lie in en o y (NFI10). Posi i e
di e ences in o al olumes on plo s wi h no ea men s,
based on da a ha could be ma ched wi h ce ain y be ween
he wo subsequen in en o ies, we e eco ded as he pai -
wise da a. The es ima ed ansi ions, he e o e, included
g ow h and mo ali y, bu no po en ial educ ions due o
calami ies o na u al dis u bances, o example.
The ansi ions due o managemen ac i i ies we e
based on simula ions o hei expec ed de elopmen . The
o es s a ec ed by inal ellings we e o ced o ansi o
he beginning o he e en-aged o a ion. Thei ea ly de el-
opmen was simula ed conse a i ely, such ha 25% o
he inal- elled a ea mo ed o olume class #2 (Appendix
1) in he i s simula ion s ep a e inal elling, and he
emainde he ea e . A hinning simula o was imple-
men ed o de i e pai wise obse a ions due o he ea -
men s. The simula o de e mined he ees o be emo ed
ollowing wo ypes o ins uc ions: (1) a hinning om
below co esponding o con en ional ins uc ions o o -
es managemen and (2) a hinning om abo e wi h an
in ensi y co esponding o an in e es a e o 3%, as p e-
dic ed by Eq.2 in Pukkala e al. (2015). As de ailed in
Vauhkonen and Packalen (2017), hese simula ions we e
applied o plo s wi h an ini ial basal a ea > 10m2/ha and
a mean heigh > 10m, which co esponds o comme cial
hinnings. In addi ion, p e-comme cial hinnings we e
simula ed o he less ma u e plo s wi h a hinning need
eco ded by he NFI. These hinnings we e implemen ed
as hinning om below, bu ins ead o applying a hinning
cu e o he ha es emo al, he aim was always o e ain
a esidual s and wi h app oxima ely 1000 ees/ha. In mim-
icking an ope a ional implemen a ion, he ees o be cu
we e dis ibu ed o di e en pa s o he diame e dis i-
bu ion, as desc ibed in de ail by Vauhkonen and Packalen
(2017). The hinnings ook place a he beginning o each
simula ion s ep. The g ow h o he o es s hinned om
below was simula ed by applying he ansi ion p obabili-
ies o o es s no managed in he simula ion s ep whe e
he hinning ook place.
The ansi ion p obabili ies we e simula ed using he
same pai wise da a o bo h age- olume and s em numbe -
olume classes and can be expec ed o de elop simila o
es ablished o es s (c . Vauhkonen and Packalen 2017).
Howe e , di e ences may occu be ween he manage-
men sys ems due o assump ions on he ea ly de elopmen
a e he egene a ion ha es . To assess he sensi i i y o
hese assump ions on he managemen sys ems, all igu es
o Sec .3 we e al e na i ely ep oduced using he same
me hodology, bu wi h di e en assump ions o he ea ly
de elopmen o e en-aged and con inuous co e o es y.
Speci ically, he pa ame e s abo e we e modi ied such ha
75% o he inal- elled a ea mo ed o he nex olume class
ha was al eady in he i s simula ion s ep a e inal ell-
ing; and he o es s hinned om abo e we e simula ed o
g ow h in he same simula ion s ep ha he ha es occu ed,
simila o o es s hinned om below, i.e. hey we e assumed
o eco e apidly despi e hea y hinning. A compa ison o
he ull esul s indica ed inc easing di e ences, i.e. unce -
ain ies owa ds he end o he simula ion pe iod. Howe e ,
he managemen sys ems we e no essen ially di e en om
each o he , based on ei he o he esul s, so ou analyses
ocused on he esul s om he conse a i e ea ly de elop-
men simula ions, while he esul s o he mo e apid de el-
opmen a e p o ided as Elec onic Supplemen a y Ma e ial
o he eade who wishes o e alua e he deg ee o sensi i -
i y in he simula ions.
Ou pu coe icien s
As he simula ions only conside ed he de elopmen o he
o es a ea dis ibu ion, sepa a e ans o ma ion coe icien s
we e de e mined o de i e u he in o ma ion o he a i-
ables o in e es (ca bon s o age, ha es emo al, and ha -
es ing cos s). The coe icien s we e de e mined as he mean
alues o he NFI plo s wi hin he dynamic classes (Appen-
dix 1) as ollows:
225Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
• Ca bon s o age The o al abo e- and below-g ound bio-
mass o each NFI plo was i s compu ed by es ima ing
s em, b anch, oliage, s ump, and oo biomass wi h he
models desc ibed in Repola (2008, 2009). The biomass
componen s we e mul iplied by species-speci ic expan-
sion ac o s (app oxima ely 0.5; see Table1 in Pukkala
2014) and summed o ob ain he ca bon con en in each
plo . The ans o ma ion coe icien s we e compu ed as
mean alues o he olume classes (Appendix 1) using
all NFI plo s.
• Ha es emo al The p opo ions o log- and pulp wood
o he o al olume ob ained om he di e en ha es s
we e i s compu ed om he es ima es o asso men ol-
umes ob ained by heo e ically bucking he ees in he
NFI plo s, ollowing he cu - o-leng h ha es ing me hod.
The o al ellings we e compu ed by summing hese p o-
po ions. Coe icien s o he hinnings we e de e mined
as mean olumes o logs o sawn wood and pulp wood
om he NFI plo s o which he hinnings we e simu-
la ed. The coe icien s o he inal ellings we e based
on simila mean alues in all NFI plo s and de e mined
using bo h age and olume classes.
• Ha es ing cos s The ime expendi u e o cu - o-leng h
logging and oadside- anspo a ion o he ees in he
NFI plo s was es ima ed using he models desc ibed in
Rummukainen e al. (1995). The models we e applied
wi h he assump ion ha one hec a e o o es ep e-
sen ed by an NFI plo was cu o he imbe asso men s
as desc ibed abo e. The models assumed highe ime
expendi u e o hinning- ypes o ha es s compa ed
o inal ellings, bu ypical e ain condi ions and wi h
no en y ime o he logging equipmen , i.e. he ime
expendi u e alues only depended on he amoun o im-
be asso men s ob ained as a esul o he di e en ypes
o ha es s. As in Vauhkonen and Pukkala (2016), he
ha es ing cos s we e compu ed wi h he assump ion ha
ope a ing a ha es e and a o wa de cos 80 €/h and
57 €/h, espec i ely. The ans o ma ion coe icien s o
ha es ing cos s (exp essed as €/m3) we e compu ed as
mean alues o he olume classes based ei he on all
NFI plo s ( inal ellings) o plo s o which he hinnings
we e simula ed ( hinnings).
Resul s
To al e ec s unde business‑as‑usual (BAU)
managemen
Assuming BAU managemen and he cu en wood a ail-
abili y ca ego ies, he o al ca bon s o age in he abo e- and
below-g ound li ing biomass was 883.2 million onnes
ca bon a he end o he simula ion pe iod. In o al, 4057.1
million m3 o oundwood was ha es ed du ing he simula-
ion (a e age: 73.8 million m3/a, wi h a a ia ion om 61.9
o 79.5 million m3/a be ween he simula ion s eps). The uni
cos s o ha es ing a ied om 16.9 o 21.5 €/m3 (a e age:
18.85 €/m3) be ween he simula ion s eps. Figu es1 and 2
show how hese alues de eloped acco ding o he a ying
p opo ions o wood a ailabili y.
All land ansi ion al e na i es esul ed in inc eased
ca bon s o age in he abo e- and below-g ound li ing bio-
mass a he end o he simula ion pe iod, compa ed o cu -
en wood a ailabili y (Fig.1, op panel). The MUCLlow
al e na i e esul ed in he highes le els o ca bon s o age
(903.8–1090.1 million onnes ca bon, depending on p%).
The highe he p% alue assigned o MUCLlow, he g ea e
he di e ence o he o he land ansi ion al e na i es. The
la e beha ed simila ly when compa ed o each o he in
e ms o he inc ease in ca bon s o age and he magni ude
o his inc emen as a unc ion o p% (Fig.1, op panel).
The educ ion in po en ial ha es emo al was g ea es
in he MUCLhigh al e na i e, ollowed by he MUCLlow
al e na i e (Fig.1, middle panel). To al ha es emo als
(compu ed o 55yea s by empo ally alloca ing he ha -
es s o he beginning o he simula ion s eps in EFDM)
a ied om 3984.5 o 3371.2 million m3 in MUCLhigh, and
om 3980.0 o 3508.6 million m3 in MUCLlow. Remo al
depended on he p%: he highe he alue, he g ea e he
di e ence o he ha es emo als o BAU managemen and
wood a ailabili y. Ha es emo als in he o he wo land-
use ansi ion al e na i es a ied om 4051 o 3887 million
m3 acco ding o p%, i.e. he di e ence o BAU managemen
was much less compa ed o he MUCLhigh and MUCLlow
al e na i es. The di e ence in ha es emo als p oduced
by he o he wo s a egies emained cons an despi e he
inc ease in p%. The MULhigh al e na i e esul ed in he
leas educ ion in ha es emo als among he conside ed
land ansi ion al e na i es.
Ha es ing cos s inc eased in conjunc ion wi h he
inc easing p opo ion o land ansi ed om FAWS, com-
pa ed o BAU managemen and wood a ailabili y. This esul
applied o all land ansi ion al e na i es. Howe e , bo h
he MUCLlow and MULlow al e na i es inc eased cos s
sligh ly, wi h uni cos s a ying be ween 19.0 and 21.15 €/
m3, depending on he p% alue. In addi ion, hese cos s we e
no a ec ed o any ex en by he inc ease in p%. On he
con a y, he e was a s ong posi i e co ela ion be ween he
p% alue and cos s in he MULhigh and MUCLhigh s a e-
gies. The la e s a egy also esul ed in he highes uni cos s
(19.9–30.2 €/m3, depending on he p% alue).
Tempo al e ec s unde BAU managemen
Ca bon s o age in he abo e- and below-g ound li ing bio-
mass, ha es emo als, and ha es ing cos s de eloped
226 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
o e ime un il he end o he simula ion pe iod (Fig.2). As
men ioned in Sec .2.2.2 and u he discussed in Sec .4,
he ha es s o he i s simula ion s ep we e i e a ed o he
same le el in all land ansi ion al e na i es, so he di e -
ences begin o show immedia ely a e he i s simula ion
s ep. Ca bon s o age s a ed o inc ease in app oxima ely
wo s eps (10yea s) a e he land ansi ions, whe eas he
inc ease in he ha es ing cos s ook place immedia ely.
Wi h espec o ca bon s o age, he land ansi ion al e na-
i es di e ed in magni ude, bu he ends we e simila o e
ime. The land-use ansi ion al e na i es di e ed mo e wi h
espec o ha es emo als and cos s.
While Fig.1 sugges s ha he inc ease in p% almos
equally a ec ed he o al ha es le els based on he MUL-
high and MULlow land ansi ions, hese al e na i es
clea ly di e ed in e ms o empo al de elopmen pa -
e n o e he en i e simula ion (Fig.2). In pa icula , he
ha es s o MULlow we e close o BAU o e he i s
ew simula ion s eps, whe eas he ha es s o MULhigh
app oached hose o BAU owa ds he end o he simu-
la ion, e en ually exceeding ha le el. MULhigh also
di e ed om he o he s a egies in ha he highes p%
alue allowed he g ea es inc eases in ha es owa ds
he end o he simula ion. The empo al de elopmen o
ha es ing cos s also di e ed depending on land ansi ion
s a egy. In al e na i es wi h emphases on high cons al,
he cos s inc eased conside ably due o he land ansi ion,
bu hei empo al de elopmen esembled BAU (i.e. he
cos s dec eased sligh ly o e ime). In al e na i es wi h
Fig. 2 Tempo al de elopmen o ca bon s ock ( op ow), ha es
emo als (middle), and ha es ing cos s (bo om), assuming di e en
deg ees o wood a ailabili y. The g ey ba s o each sub- igu e depic
he de elopmen unde business-as-usual ac i i y and ansi ion p ob-
abili ies and cu en wood a ailabili y. The eigh lines o each sub-
igu e deno e shi s o p% (p = 5, 10, …, o 40; see Sec .2.1) o he
land a ea o Fo es s A ailable o Wood Supply (FAWS) o Fo es s
wi h Res ic ions on A ailabili y o Wood Supply (FRAWS) o Fo -
es s No A ailable o Wood Supply (FNAWS) acco ding o he ou
land ansi ion s a egies
227Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
emphases on low cons al, he cos s did no ini ially di e
om BAU, bu ins ead inc eased o e ime.
Join e ec s due oshi s be weenwood a ailabili y
ca ego ies andchanges in o al ha es ing le els
andalloca ion
Al e ing ha es le els and hei alloca ions a ec ed he le -
els o ca bon s o age in he abo e- and below-g ound li ing
biomass, ha es emo als, and ha es ing cos s unde cu -
en wood a ailabili y, as illus a ed in Fig.3 in a scale no -
malized o he le el o he ini ial alues o hese a ibu es
assuming BAU managemen and cu en wood a ailabili y
(g ey ba s in he le mos column o Fig.3). Rela i e o ha
si ua ion, mo e ha es emo als we e ob ained a lowe
cos s when he ha es alloca ion was changed om ABAU
o ASB. This change also inc eased ca bon s o age owa ds
he end o he simula ion. Ca bon s o age dec eased and
ha es ing cos s inc eased when he ABAU ha es alloca-
ion was main ained, bu he p opo ion o ha es ed a eas
was inc eased a he beginning o he simula ion o co e-
spond wi h he ha es goals ou lined in he Na ional Fo es
S a egy. Howe e , he alloca ion o inc eased ha es le els
(acco ding o ASB) somewha compensa ed o hese e ec s
Fig. 3 The empo al de elopmen o ca bon s ock ( op ow), ha es
emo als (middle), and ha es ing cos s (bo om), assuming di e en
deg ees o wood a ailabili y and ha es ing. The y-axes a e p esen ed
in a scale no malized o he le el o he ini ial alues o hese a ib-
u es assuming business-as-usual ac i i y and ansi ion p obabili ies,
he de elopmen o which is illus a ed in sub- igu e “ABAU; 62 mill”
acco ding o he applied alloca ion o ha es s and he o al amoun
o ha es s in m3/a o he i s simula ion s ep. The g ey ba s depic
he de elopmen assuming cu en wood a ailabili y, while he col-
ou ed lines indica e p = 25% o he land a ea o Fo es s A ailable o
Wood Supply (FAWS) shi ed o Fo es s wi h Res ic ions on A ail-
abili y o Wood Supply (FRAWS) o Fo es s No A ailable o Wood
Supply (FNAWS) acco ding o he di e en s a egies ( e e o Fig.1
cap ion o he in e p e a ion o he symbols). The black lines indica e
he hypo he ical si ua ion whe e all o es s a e conside ed as FAWS
and simula ed acco ding o he e en-aged managemen sys em
234 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
Appendix2
The pa ame e iza ion o he EFDM equi es ha NFI da a
a e classi ied in o ac o s ha ep esen he ini ial o es
s a e. T ansi ion p obabili ies, managemen ac i i ies, and
ou pu coe icien s a e speci ied o each ac o combina ion
(Vauhkonen and Packalen 2017). The p esen s udy di e s
in ha mo e han one ini ial s a e was de ined o he simula-
ions by ans e ing di e en p opo ions o FAWS o he
o he wood a ailabili y ca ego ies acco ding o a conse -
a ion alue p oxy (c . Sec .2.1). This appendix aims o
(nume ically) exempli y he compu a ions ela ed o he con-
se a ion alue (1), classi ica ion o he NFI da a o ob ain
he ini ial s a e (2) and unning he EFDM simula ions (3)
on an example o es . Conside ha an NFI plo wi hou o -
es use es ic ions was loca ed on p i a e land in sou he n
Finland. In o al, 19 ees o sp uce, bi ch, and aspen species
ha we e g owing on he mos e ile soil we e measu ed,
wi h he b eas heigh diame e s (d; measu emen s in mm)
dis ibu ed as ollows: dsp uce = {323, 265, 219, 249, 238,
427, 291, 345, 313, 407, 445, 505, 407, 402}; dbi ch = {231,
199, 177, 147}; daspen = {414}.
Conse a ion alue p oxy
The p opo ion o land ans e ed om FAWS o he o he
ca ego ies was selec ed acco ding o a conse a ion alue
p oxy (adap ed omLeh omäki e al. 2015):
whe e diame e and olume a e he species-speci ic mean
diame e and g owing s ock olume o a plo , () is a species-
speci ic ans o ma ion unc ion used o con e he diame e
o a conse a ion alue index based on expe knowledge,
w is a weigh ing ha co esponds o si e e ili y, and sp is
a species index. To compu e cons al, he ees measu ed
om each NFI plo we e assigned o g oups o pine, sp uce,
bi ch, and o he deciduous species. The median and maxi-
mum alues o he mean diame e s o hese species we e
(1)
cons al
=
∑
sp
[wsp × (diame e sp)× olumesp]
,
compu ed o each o he 15 o es y egions in Finland.
Species-speci ic asymp o e and scaling pa ame e s (speci i-
cally, mod_asym, and pa ame e s wi h su ices la dia and
a dia om he ile pa ame e s-esmk.cs ; Leh omäki 2015)
we e ela ed o he median and maximum alues (as ca -
ied ou in ile gis.calcula e.index.R; Leh omäki 2015). The
pa ame e s p oduced sigmoidal ans o ma ion unc ions
(Fig.6) ha we e used o ans o m he species-speci ic
mean diame e s o conse a ion indices be ween 0 and 1.
Figu e6 shows he species-speci ic mean diame e s o he
example plo , he esul ing conse a ion index alues and
sigmoidal ans o ma ion unc ions o e e y o es y egion
and species. Smalle diame e s yield g ea e ans o med
alues in deciduous species han in coni e ous species. Fo
bi ch, his ans o ma ion was s ongly o es y egion spe-
ci ic, in ha occu ences o la ge bi ch ees yielded a highe
conse a ion alue in egions whe e he median diame e o
bi ch is low.
The same species-speci ic weigh ings o si e e ili y,
as used by Leh omäki e al. (2015) in hei “Coa se wi h
classes” wo k low o conse a ion p io i iza ion based on
mul i-sou ce NFI da a, we e ex ac ed. The example plo
abo e was assigned si e weigh ings o 3.0, 4.0, and 7.0 o
sp uce, bi ch and aspen, espec i ely, as i was loca ed on
a Uni s in m3/ha o olume, 1/ha o s em numbe , and yea s o age
Table 1 (con inued)
Age classesa
21. (95, 100]
22. (100, 105]
23. (105, 110]
24. (110, 115]
25. (115, 120]
26. (120, ∞)
Fig. 6 Sigmoidal ans o ma ions om species-speci ic mean diam-
e e o conse a ion alue. Black, ed, blue, and g een lines show he
ans o ma ion unc ions o pine, sp uce, bi ch, and o he deciduous
ees, espec i ely, and he colou ed do s show how he ees meas-
u ed om he example plo ela ed o he ans o ma ion unc ions.
Sepa a e lines a e d awn o depic he di e en o es y egions. No e
ha o some species, he ans o ma ion is de e mined piecewise,
wi h di e en asymp o e and scaling pa ame e s applied o diam-
e e s below and abo e he egion-speci ic median alue. (Colo igu e
online)

235Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
highly e ile soil. To ob ain he inal cons al o he plo ,
he ans o med diame e s we e mul iplied by species-spe-
ci ic o al s em olume and weigh ing o si e e ili y and
summed o e he species in a plo . The o al cons al alue
(907.12) o he example plo was hus ob ained as a sum
o Eq.1 applied pe species as ollows:
Acco ding o he example abo e, he conse a ion alue
would be lowe o plo s wi h he same olume bu wi h
less species. Si e weigh ings a ec he esul conside ably;
he weigh ings a e educed o 1.0–1.5 o coni e species
on less e ile soils. The weigh ings a e also educed o
deciduous species, al hough less se e ely, and he occu -
ence o deciduous species in he leas e ile soils a e
also weigh ed sligh ly highe . O e all, he cons al index
was ound o e lec well he g owing s ock, species and
si e a ia ion, ela i e o a ia ion wi hin o es y egions
(Fig.6).
(2)
cons alsp uce
=3.0 ×0.78 ×310.38 m
3
∕ha ≈
725.8
(3)
cons albi ch
=4.0 ×0.02 ×60.734 m
3
∕ha ≈
4.9
(4)
cons alaspen
=7.0 ×1.0 ×25.212 m
3
∕ha ≈
176.5
EFDM inpu da a
The example plo ep esen s an a ea o 350ha when compu ed
acco ding o NFI me hodology o o es a ea es ima ion (c .
Vauhkonen and Packalen 2017). I is classi ied in he ini ial
o es a ea dis ibu ion ma ix as a sp uce-domina ed, p i a e
FAWS in he highes axa ion class acco ding o he s a ic ac o s
(Sec .2.2.1). I is u he classi ied o olume class #12, s em
numbe class #6, and age class #17 acco ding o i s o al ol-
ume (396m3/ha), s em numbe (730 s ems/ha), age (78yea s),
and class limi s (Appendix 1). A simila classi ica ion is applied
o each plo in bo h he ull and pai wise da a, and hese da a
sou ces a e used o de i e he ini ial s a e, ac i i y/ ansi ion
p obabili ies, and ou pu coe icien s as desc ibed in Sec .2.2
and in mo e de ail by Vauhkonen and Packalen (2017).
EFDM simula ions
The o es a ea dis ibu ion a e one simula ion s ep is
ob ained by wo ma ix mul iplica ions (see Sec .3 in Pack-
alen e al. 2014): (1) he cu en a ea dis ibu ion is mul-
iplied by he ac i i y p obabili ies, which yields sepa a e
ma ices ha ep esen he a ea o each cell a ec ed by each
ac i i y; and (2) he a o emen ioned in e media e ma ices
a e mul iplied by he ansi ion p obabili ies, which yields
he a ea dis ibu ion in he nex s ep. Figu e7 depic s he
Fig. 7 The p opo ion o
di e en managemen ac i i-
ies applied o he example
plo on he i s s ep o he
simula ions. The ac i i y p ob-
abili ies a y depending on
he choice be ween business-
as-usual alloca ion (ABAU) o
schoolbook-alloca ion (ASB) and
whe he he o es is conside ed
as Fo es s A ailable o Wood
Supply (FAWS) o Fo es s wi h
Res ic ions on A ailabili y o
Wood Supply (FRAWS) and,
he e o e, simula ed using age-
olume o s em numbe - olume
ma ices, espec i ely
236 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
ac i i y p obabili ies ha a e applied o he example plo ,
depending on whe he he plo is conside ed ei he as FAWS
o FRAWS and managed acco ding o ei he ABAU o ASB.
Finally, he p opo ions a e adjus ed by wo al e na i e o al
ha es ing amoun s, bu hese e ec s a e omi ed he e o
cla i y pu poses.
The a ea managed by hinnings o inal ellings p oduces
ha es ed emo als. Acco ding o he ou pu coe icien s
o he plo in he i s simula ion s ep, inal elling o he
plo desc ibed abo e would yield in o al 200.5m3/ha and
162.8m3/ha o logs o sawn wood and pulp wood, espec-
i ely, a an es ima ed ( o al) cos o 8.76 €/m3. The co e-
sponding alues a e 57.8m3/ha and 89.1m3/ha o hinning
om below ( o al cos o 15.6 €/m3) and 129.9m3/ha and
95.8m3/ha ( o al cos 12.3 €/m3) o hinning om abo e.
The ansi ion p obabili ies shi he ini ial a ea o he plo
o mul iple ma ix cells. Using age- olume classes, he o es
ha is no managed gains age, bu emains in he (highes )
olume class. Thinnings om below ansi he a ea o ol-
ume classes om #7 upwa ds, he mos common class being
#10. Using s em numbe , abou hal o he olume emains
in he same s em numbe and olume class. Ha es s shi
he a ea om s em numbe / olume classes #3/#6 upwa ds,
he mos common a ge class being #4/#7. No ably, na u al
p ocesses wi h s em numbe may shi he a ea bo h up and
down in e ms o classes, and shi he a ea downwa ds om
bo h hinning om below and abo e, al hough based on di -
e en hinning ules. In he nex simula ion s ep, he o es
a ea o he ecei ing classes is u he upda ed using he
ac i i y and ansi ion p obabili ies o he speci ic classes.
Figu e8 shows how he ini ial a ea o 350ha in one class
e ol ed du ing he 10 simula ion s eps based on ei he age-
olume and s em numbe - olume classes. Because o he
mul iple ansi ions ha ake place due o ac i i y and ansi-
ion p obabili ies, i migh no be easible o ack he de el-
opmen o o es s in a single class, bu he o e all de elop-
men o he class s uc u e is deemed ealis ic wi h espec
o he p ope ies o he o es .
Re e ences
Albe di I, Michalak R, Fische C, Gaspa ini P, B ändli UB, Tom e
SM, Kuliesis A, Sno ason A, Redmond J, He nández L, Lanz A,
Vidondo B, S oyano N, S oyano a M, Ves man M, Ba ei o S,
Ma in G, Cañellas I, Vidal C (2016) Towa ds ha monized assess-
men o Eu opean o es a ailabili y o wood supply in Eu ope.
Fo Pol Econ 70:20–29
Al ahahleh L, Ikonen VP, Kilpeläinen A, To ssonen P, S andman H,
Asikainen A, Kau ola J, Venäläinen A, Pel ola H (2016) E ec s
o o es conse a ion and managemen on olume g ow h,
Fig. 8 Dis ibu ion o age- ol-
ume (abo e) and s em numbe -
olume (below) classes a e
simula ing 10 ime-s eps and
applying class-speci ic ac i i y
and ansi ion p obabili ies o
he ini ial 350ha o o es . Re e
o Appendix 1 o he in e p e-
a ion o class numbe s on he
x-axes. The g ey ones wi hin
he ba s depic olume classes,
he da kes g ey e e ing o he
lowes numbe o olume class
and ice e sa
237Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
ha es ed amoun o imbe , ca bon s ock, and amoun o dead-
wood in Finnish bo eal o es s unde changing clima e. Can J Fo
Res 47:215–225
A ponen A, Leh omäki J, Leppänen J, Tomppo E, Moilanen A (2012)
E ec s o connec i i y and spa ial esolu ion o analyses on
conse a ion p io i iza ion ac oss la ge ex en s. Conse Biol
26:294–304
Bibe P e al (2015) How sensi i e a e ecosys em se ices in Eu opean
o es landscapes o sil icul u al ea men ? Fo es s 6:1666–1695
Bollandsås OM, Buongio no J, Gobakken T (2008) P edic ing he
g ow h o s ands o ees o mixed species and size: a ma ix
model o No way. Scand J Fo Res 23:167–178
Bose AK, Weiski el A, Kuehne C, Wagne RG, Tu nblom E, Bu kha
HE (2018) Does comme cial hinning imp o e s and-le el g ow h
o he h ee mos comme cially impo an so wood o es ypes
in No h Ame ica? Fo Ecol Manag 409:683–693
Ca pen ie S, Filo as E, Handa IT, Messie C (2017) T ade-o s
be ween imbe p oduc ion, ca bon s ocking and habi a quali y
when managing woodlo s o mul iple ecosys em se ices. En i-
on Conse 44:14–23
CBD (2010) Con en ion on biological di e si y: COP 10 decision X/2.
S a egic plan o biodi e si y 2011–2020. h ps ://www.cbd.in /
decis ion/cop/?id=12268 . Accessed 27 Dec 2017
C eu zbu g MK, Schelle RM, Lucash MS, LeDuc SD, Johnson MG
(2017) Fo es managemen scena io in a changing clima e: ade-
o s be ween ca bon, imbe , and old o es . Ecol Appl 27:503–518
Diaci J, Ke G, O’Ha a K (2011) Twen y- i s cen u y o es y:
in eg a ing ecologically based, une en-aged sil icul u e wi h
inc eased demands on o es s. Fo es y 84:463–465
Diaz-Bal ei o L, Alonso R, Ma ínez-Jaú egui M, Pa dos M (2017)
Selec ing he bes o es managemen al e na i e by agg ega ing
ecosys em se ices indica o s o e ime: a case s udy in cen al
Spain. Ecol Indic 72:322–329
Fü s enau C, Badek FW, Lasch P, Lexe MJ, Linde M, Moh P,
Suckow F (2007) Mul iple-use o es managemen in conside a-
ion o clima e change and he in e es s o s akeholde g oups. Eu
J Fo Res 126:225–239
G assi G, den Elzen MG, Ho AF, Pilli R, Fede ici S (2012) The ole
o he land use, land use change and o es y sec o in achie ing
Annex I educ ion pledges. Clim Change 115:873–881
Hanski I (2011) Habi a loss, he dynamics o biodi e si y, and a pe -
spec i e on conse a ion. Ambio 40:248–255
Heinonen T, Pukkala T, Meh ä alo L, Asikainen A, Kangas J, Pel ola
H (2017) Scena io analyses o he e ec s o ha es ing in ensi y
on de elopmen o o es esou ces, imbe supply, ca bon bal-
ance and biodi e si y o Finnish o es y. Fo Pol Econ 80:80–98
Heinonen T, Pukkala T, Asikainen A, Pel ola H (2018) Scena io analy-
ses on he e ec s o e iliza ion, imp o ed egene a ion ma e ial,
and di ch ne wo k main enance on imbe p oduc ion o Finnish
o es s. Eu J Fo Res 137:93–107
Hynynen J, Salminen H, Ah ikoski A, Huuskonen S, Ojansuu R, Siip-
ileh o J, Leh onen M, Ee ikäinen K (2015) Long- e m impac s
o o es managemen on biomass supply and o es esou ce
de elopmen : a scena io analysis o Finland. Eu J Fo Res
134:415–431
Kä kkäinen L, Haakana H, Hi elä H, Packalen T (2018) Using a
decision suppo sys em o s udy impac s o land use policies on
wood p ocu emen possibili ies o he sawmill indus y—a case
s udy a egional and municipal le els. Pol Econ, Fo . h ps ://doi.
o g/10.1016/j. o po l.2017.10.002
Knoke T (2012) The economics o con inuous co e o es y. In: Puk-
kala T, on Gadow K (eds) Con inuous co e o es y, managing
o es ecosys ems 23. Sp inge , Do d ech , pp 167–193
Ko iaho JS, Kuusela S, Nieminen E, Päi inen J, Moilanen A (2016)
F amewo k o assessing and e e sing ecosys em deg ada ion—
epo o he Finnish es o a ion p io i iza ion wo king g oup on
he op ions and cos s o mee ing he Aichi biodi e si y a ge o
es o ing a leas 15 pe cen o deg aded ecosys ems in Finland.
Repo s o he Minis y o En i onmen 15en, Ympä is öminis-
e iö, Helsinki. h p://julka isu . al i oneu os o. i/handl e/10024
/74862 . Accessed 27 Dec 2017
Kuulu ainen T, Tah onen O, Aakala T (2012) E en-aged and une en-
aged o es managemen in bo eal Fennoscandia: a e iew. Ambio
41:720–737
La ond V, Co donnie T, Mao Z, Cou baud B (2017) T ade-o s
be ween ecosys em se ices in une en-aged moun ain o es s:
e idences using Pa e o on s. Eu J Fo Res 136:997–1012
Leh omäki J (2015) zse up-esmk: PONE publica ion (Ve sion 1.1).
Zenodo ile eposi o y. h ps ://doi.o g/10.5281/zenod o.27240 .
Accessed 22 Oc 2018
Leh omäki J, Tomppo E, Kuokkanen P, Hanski I, Moilanen A (2009)
Applying spa ial conse a ion p io i iza ion so wa e and high-
esolu ion GIS da a o a na ional-scale s udy in o es conse a-
ion. Fo Ecol Manag 258:2439–2449
Leh omäki J, Tuominen S, Toi onen T, Leinonen A (2015) Wha
da a o use o o es conse a ion planning? A compa ison o
coa se open and de ailed p op ie a y o es in en o y da a in Fin-
land. PLoS ONE 10(8):e0135926. h ps ://doi.o g/10.1371/jou n
al.pone.01359 26
Lunds öm J, Öhman K, Rönnq is M, Gus a sson L (2016) Con-
side ing u u e po en ial ega ding s uc u al di e si y in selec-
ion o o es ese es. PLoS ONE 11(2):e0148960. h ps ://doi.
o g/10.1371/jou n al.pone.01489 60
MAF (2015) Minis y o Ag icul u e and Fo es y: Na ional Fo es
S a egy 2025—Go e nmen Resolu ion o 12 Feb ua y 2015,
Minis y o Ag icul u e and Fo es y, Helsinki. h p://mmm. i/
en/n s. Accessed 27 Dec 2017
Meh ä alo L, Pel ola H, Kilpeläinen A, Ikonen VP (2014) The esponse
o basal a ea g ow h o Sco s pine o hinning: a longi udinal anal-
ysis o ee-speci ic se ies using a nonlinea mixed-e ec s model.
Fo Sci 60:636–644
Miina J, Pukkala T, Ku ila M (2016) Op imal mul i-p oduc manage-
men o s ands p oducing imbe and wild be ies. Eu J Fo Res
135:781–794
Moilanen A, Ande son BJ, Eigenb od F, Heinemeye A, Roy DB, Gill-
ings S, A mswo h PR, Gas on KJ, Thomas CD (2011) Balanc-
ing al e na i e land uses in conse a ion p io i iza ion. Ecol Appl
21:1419–1426
Mon o o Gi ona M, Rossi S, Lussie JM, Walsh D, Mo in H (2017)
Unde s anding ee g ow h esponses a e pa ial cu ings: a new
app oach. PLoS ONE 12(2):e0172653. h ps ://doi.o g/10.1371/
jou n al.pone.01726 53
Mouche MA, Rega C, Lasseu R, Geo ges D, Pa acchini ML, Renaud
J, S ü ck J, Schulp CJE, Ve bu g PH, Ve ke k PJ, La o el S (2017)
Ecosys em se ice supply by Eu opean landscapes unde al e na-
i e land-use and en i onmen al policies. In J Biodi Sci Ecosys
Se Manag 13:342–354
Neumann M, Mo eno A, Mues V, Hä könen S, Mu a M, Bou iaud O,
Lang M, Ach en WMJ, Thi olle-Caza A, B onisz K, Me ganič J,
Decuype M, Albe di I, As up R, Moh en F, Hasenaue H (2016)
Compa ison o ca bon es ima ion me hods o Eu opean o es s.
Fo Ecol Manag 361:397–420
Ne alainen S (2017) Compa ison o damage isks in e en-and une en-
aged o es y in Finland. Sil a Fenn. h ps ://doi.o g/10.14214 /
s .1741
Nieminen M, Hökkä H, Laiho R, Juu inen A, Ah ikoski A, Pea son M,
Kojola S, Sa kkola S, Launiainen S, Valkonen S, Pen ilä T, Lohila
A, Saa inen M, Haah i K, Mäkipää R, Mie inen J, Ollikainen M
(2018) Could con inuous co e o es y be an economically and
en i onmen ally easible managemen op ion on d ained bo eal
pea lands? Fo Ecol Manag 424:78–84
238 Eu opean Jou nal o Fo es Resea ch (2019) 138:219–238
1 3
Packalen T, Sallnäs O, Si kiä S, Ko honen K, Salminen O, Vidal C,
Robe N, Colin A, Beloua d T, Schadaue K, Be ge A, Rego A,
Lou o G, Camia A, Rä y M, San-Miguel J (2014) The Eu opean
o es y dynamics model: concep , design and esul s o i s case
s udies. JRC Science and Policy Repo s Volume 93450, EUR
27004. Publica ions O ice o he Eu opean Union, Luxembou g.
h ps ://doi.o g/10.2788/15399 0
Pang X, No ds öm EM, Bö che H, T ubins R, Mö be g U (2017)
T ade-o s and syne gies among ecosys em se ices unde di e -
en o es managemen scena ios— he LEcA ool. Ecosys Se
28:67–79
Peu a M, Bu gas D, Ey indson K, Repo A, Mönkkönen M (2018)
Con inuous co e o es y is a cos -e icien ool o inc ease mul-
i unc ionali y o bo eal p oduc ion o es s in Fennoscandia. Biol
Conse 217:104–112
Pue mann KJ, Wilson SM, Bake SC, Donoso PJ, D össle L, Amen e
G, Ha ey BD, Knoke T, Lu Y, Nocen ini S, Pu z FE, Yoshida T,
Bauhus J (2015) Sil icul u al al e na i es o con en ional e en-
aged o es managemen —Wha limi s global adop ion? Fo Eco-
sys 2:8. h ps ://doi.o g/10.1186/s4066 3-015-0031-x
Pukkala T (2014) Does bio uel ha es ing and con inuous co e man-
agemen inc ease ca bon seques a ion? Fo Pol Econ 43:41–50
Pukkala T (2016a) Plen e wald, Daue wald, o clea cu ? Fo Pol Econ
62:125–134
Pukkala T (2016b) Which ype o o es managemen p o ides mos
ecosys em se ices? Fo Ecosys 3:9. h ps ://doi.o g/10.1186/
s4066 3-016-0068-5
Pukkala T, Kellomäki S, Mus onen E (1988) P edic ion o he ameni y
o a ee s and. Scand J Fo Res 3:533–544
Pukkala T, Lähde E, Laiho O, Salo K, Ho anen JP (2011) A mul i unc-
ional compa ison o e en-aged and une en-aged o es manage-
men in a bo eal egion. Can J Fo Res 41:851–862
Pukkala T, Lähde E, Laiho O (2013) Species in e ac ions in he
dynamics o e en-and une en-aged bo eal o es s. J Sus ain Fo
32:371–403
Pukkala T, Lähde E, Laiho O (2015) Which ees should be emo ed
in hinning ea men s? Fo Ecosys 2:32. h ps ://doi.o g/10.1186/
s4066 3-015-0056-1
R Co e Team (2016) R: a language and en i onmen o s a is ical com-
pu ing. R ounda ion o s a is ical compu ing, Vienna, Aus ia.
h ps ://www.R-p oje c .o g/. Accessed 6 Oc 2017
Repola J (2008) Biomass equa ions o bi ch in Finland. Sil a Fenn
42:605–624
Repola J (2009) Biomass equa ions o Sco s pine and No way sp uce
in Finland. Sil a Fenn 43:625–647
Roessige J, Ficko A, Clasen C, G iess VC, Knoke T (2016) Va iabili y
in g ow h o ees in une en-aged s ands displays he need o
op imizing di e si ied ha es diame e s. Eu J Fo 135:283–295
Rummukainen A, Alanne H, Mikkonen E (1995) Wood p ocu emen in
he p essu e o change: esou ce e alua ion model ill yea 2010.
Ac a Fo Fenn 248:1–98
Rybicki J, Hanski I (2013) Species–a ea ela ionships and ex inc ions
caused by habi a loss and agmen a ion. Ecol Le 16(S1):27–38
Schou E, Meilby H (2013) T ans o ma ion o e en-aged Eu opean
beech (Fagus syl a ica L.) o une en-aged managemen unde
changing g ow h condi ions caused by clima e change. Eu J Fo
Res 132:777–789
Sch ö e M, Rusch GM, Ba on DN, Blumen a h S, No dén B (2014)
Ecosys em se ices and oppo uni y cos s shi spa ial p io i ies
o conse ing o es biodi e si y. PLoS ONE 9(11):e112557.
h ps ://doi.o g/10.1371/jou n al.pone.01125 57
Si kiä S (2012) Me hodology and sys em design—Appendix1 in
Anon. De eloping and es ing a p o o ype o Eu opean o es y
dynamics model (EFDM), F amewo k con ac o he p o i-
sion o o es da a and se ices in suppo o he Eu opean Fo -
es Da a Cen e. Speci ic Con ac 10 Repo , Re e ence: 2007/S
194-235358 o 09/10/2007. h ps ://gi hu b.com/ec-j c/e dm/blob/
mas e /docum en s/EFDMi ns u c ion s/Seija _Ma he ma ic s_behin
d_EFDM.pd Accessed 14 Dec 2018
Si kiä S, Leh omäki J, Lindén H, Tomppo E, Moilanen A (2012)
De ining spa ial p io i ies o cape caillie Te ao u ogallus lek-
king landscape conse a ion in sou h-cen al Finland. Wildl Biol
18:337–353
Solbe g B, Be gseng E, Linds ad BH (2017) Assessing na ional impac s
o in e na ional en i onmen al egimes o biodi e si y p o ec ion
and clima e mi iga io in bo eal o es y—expe iences om using
a quan i a i e app oach. Fo Pol Econ 85:147–160
S e ens J, Mon gome y C (2002) Unde s anding he compa ibili y o
mul iple uses on o es land—a su ey o mul i esou ce esea ch
wi h applica ion o he Paci ic No hwes . U.S. Depa men o
Ag icul u e, Fo es Se ice, Paci ic No hwes Resea ch S a ion.
h ps ://www. s.usda.go / ees ea ch /pubs/4393. Accessed 25 Jan
2018
S ands öm M (2017) Timbe ha es ing and long-dis ance anspo a-
ion o oundwood 2016. In: Me sä ehon uloskal osa ja 1b/2017,
h p://www.me sa eho. i/ imbe -ha e s ing -and-long-dis a nce-
ans po a ion-o - ound wood-2016/. Accessed 25 Jan 2018
S engbom J, Axelsson EP, Lundma k T, No din A (2018) T ade-o s
in he mul i-use po en ial o managed bo eal o es s. J Appl Ecol
55:958–966
Su he land IJ, Benne EM, Ge gel SE (2016) Reco e y ends o mul-
iple ecosys em se ices e eal non-linea esponses and long-
e m adeo s om empe a e o es ha es ing. Fo Ecol Manag
374:61–70
T i iño M, Juu inen A, Mazzio a A, Mie inen K, Podkopae D,
Reunanen P, Mönkkönen M (2015) Managing a bo eal o es
landscape o p o iding imbe , s o ing and seques e ing ca bon.
Ecosys Se 14:179–189
an de Plas F e al (2018) Con inen al mapping o o es ecosys em
unc ions e eals a high bu un ealised po en ial o o es mul i-
unc ionali y. Ecol Le 21:31–42
Vauhkonen J, Packalen T (2017) A Ma ko chain model o simu-
la ing wood supply om any-aged o es managemen based on
Na ional Fo es In en o y (NFI) da a. Fo es s 8(9):307. h ps ://
doi.o g/10.3390/ 8090 307
Vauhkonen J, Packalen T (2018) Unce ain ies ela ed o clima e
change and o es managemen wi h implica ions on clima e egu-
la ion in Finland. Ecosys Se 33(B):213–224
Vauhkonen J, Pukkala T (2016) Selec ing he ees o be ha es ed
based on he ela i e alue g ow h o he emaining ees. Eu J
Fo Res 135:581–592
Vauhkonen J, Ruo salainen R (2017) Assessing he p o isioning
po en ial o ecosys em se ices in a Scandina ian bo eal o es :
sui abili y and adeo analyses on g id-based wall- o-wall o es
in en o y da a. Fo Ecol Manag 389:272–284
Ve ke k PJ, Ma sa R, Gie giczny M, Lindne M, Edwa ds D, Schel-
haas MJ (2014) Assessing impac s o in ensi ied biomass p oduc-
ion and biodi e si y p o ec ion on ecosys em se ices p o ided
by Eu opean o es s. Ecosys Se 9:155–165
Y jölä T (2002) Fo es managemen guidelines and p ac ices in Fin-
land, Sweden and No way. In e nal Repo 11, Eu opean Fo es
Ins i u e, Joensuu. h p://www.e i.in / iles /a ac hmen s/publi ca io
ns/i _11.pd . Accessed 27 Dec 2017