Name o he Deli e able
Quan i ied ex-an e
global impac s om
selec ed supply chains
on biodi e si y and
o he ecosys em
se ices
Deli e able 7.2
2
Summa y
Wo k Package 7
Deli e able No 7.2
Dissemina ion Le el Public
Type Repo
Lead Pa ne IIASA
Due Da e July 15 h, 2025
Submission Da e July 17 h, 2025
S a us D a
Au ho s
Da id Leclè e, Pekka Lau i, Be ina Meinha ,
Amanda Palazzo, Cla a F ezal, Ful io di Ful io, Niklas
Hinkel, Sibylle Roue -Pollakis, And ey Lessa De ci
Augus ynczik
Abou CLEVER
P ojec Numbe
101060765
P ojec Ti le
CLEVER: C ea ing le e age o enhance biodi e si y
ou comes o global biomass ade
Topic HORIZON-CL6-2021-BIODIV-01-15
S a da e
1s Sep embe 2022
End da e
31s Augus 2025
Coo dina o
RHEINISCHE FRIEDRICH-WILHELMS-UNIVERSITAT BONN
This documen has been p epa ed in he amewo k o he p ojec CLEVER. Views
and opinions exp essed a e howe e hose o he au ho (s) only and do no
necessa ily e lec hose o he Eu opean Union o he Eu opean Resea ch
Execu i e Agency. Nei he he Eu opean Union no he g an ing au ho i y can be
held esponsible o hem.
3
This p ojec has ecei ed unding om he Eu opean Resea ch
Execu i e Agency unde HORIZON Resea ch and Inno a ion Ac ions,
g an ag eemen no101060765; and om he UK Resea ch and
Inno a ion (UKRI) unde he UK go e nmen 's Ho izon Eu ope unding
gua an ee [g an numbe 10038491]
4
Con en s
Table o Con en s
EXECUTIVE SUMMARY ....................................................................................................... 6
O e iew o me hods ........................................................................................................ 6
Scena io ou comes o soy supply chains ........................................................................... 6
Scena io ou comes o o es y supply chains .................................................................... 8
Scena io ou comes o c op aqua eed supply chains .......................................................... 9
Nex s eps ....................................................................................................................... 10
INTRODUCTION & OBJECTIVES ....................................................................................... 11
Soy supply chains ............................................................................................................ 11
Fo es supply chains ........................................................................................................ 13
Aquacul u e and aqua eed supply chains ........................................................................ 14
METHODS .......................................................................................................................... 16
Soy supply chains ............................................................................................................ 16
O e all app oach ............................................................................................................................... 16
Scena io desc ip ion ......................................................................................................................... 16
Quan i ica ion ................................................................................................................................... 18
Fo es supply chains ........................................................................................................ 19
O e all app oach ............................................................................................................................... 19
Scena io desc ip ion ......................................................................................................................... 21
Quan i ica ion ................................................................................................................................... 25
Aquacul u e and aqua eed supply chains ........................................................................ 26
O e all app oach ............................................................................................................................... 26
Scena io desc ip ion ......................................................................................................................... 27
Quan i ica ion ................................................................................................................................... 30
RESULTS AND DISCUSSION .............................................................................................. 31
Soy supply chains ............................................................................................................ 31
Resul s ............................................................................................................................................... 31
Discussion ......................................................................................................................................... 54
Fo es supply chains ........................................................................................................ 59
Resul s ............................................................................................................................................... 59
Discussion ......................................................................................................................................... 71
Aquacul u e & aqua eed supply chains ............................................................................ 72
Resul s ............................................................................................................................................... 72
5
Discussion ......................................................................................................................................... 78
PROJECT OUTPUTS ACHIEVED ........................................................................................ 81
REFERENCES ...................................................................................................................... 82
APPENDIX .......................................................................................................................... 89
S akeholde wo kshop o soy supply chain scena ios ..................................................... 89
S akeholde wo kshop o o es y supply chain scena ios .............................................. 90
S akeholde wo kshop o aquacul u e & aqua eed supply chain scena ios ..................... 91
Model imp o emen s o soy supply chains ..................................................................... 92
Implemen a ion o B azilian en i onmen al p o ec ion egula ions and land-use es ic ions in
GLOBIOM .......................................................................................................................................... 92
Imp o emen o soy p oduc ion sys em dynamics ........................................................................... 93
Model imp o emen s o o es supply chains ................................................................. 97
Age-class dynamics and o es managemen ................................................................................... 97
HWP and BECCS ca bon accoun ing ................................................................................................. 99
De o es a ion and a o es a ion ..................................................................................................... 100
O he na u al land di ision o abandoned land and na u al land................................................... 101
Wood-based p oduc s ade and egional compe i i eness ........................................................... 102
Model imp o emen s o aquacul u e and aqua eed supply chains ............................... 103
P ojec ing he capaci y o un ed aquacul u e beyond 2020 ........................................................... 103
Sensi i i y analysis o he c op composi ion o c op aqua eed equi emen s ................................ 103
6
EXECUTIVE SUMMARY
The CLEVER p ojec se s ou o explo e he e ec i eness o inno a i e in e na ional ade and
supply chain go e nance in e en ions on non- ood biomass supply chains and biodi e si y
h ough he use o models and scena ios. This deli e able cons i u es he i s o wo scena io
quan i ica ion deli e ables (D7.2-3), ocused on explo a i e scena ios o h ee non-biomass
supply chains (soy, o es y and c op-based aqua eeds), as pa o he Task 7.2.
O e iew o me hods
The scena ios ha e been co-designed h ough li e a u e e iew and dedica ed s akeholde
wo kshops, and quan i ied wi h he GLOBIOM land use model, which p ojec s he dynamics o
mos impo an ag icul u al and o es y supply chains and hei socio-economic and
en i onmen al impac s om he yea 2000 in o he u u e wi h a decadal ime s ep (up o 2050
o e en 2100). Impac s a e analysed o a ious indica o s ela ed o ma ke balance (demand,
supply, ade) as well as en i onmen al (e.g., land and wa e use, GHG emissions, biodi e si y)
and economic (e.g., ood a ailabili y, alue o p oduc ion) ou comes, a global scale wi h a ocus
on he EU and B azil. A pa icula e o is made o quan i y impac s on biodi e si y, wi h mul iple
ecosys ems and impac pa hways conside ed when possible.
A b ie in oduc ion o he modelling amewo k is p o ided in he deli e able D7.1, while a
de ailed desc ip ion o dedica ed modelling imp o emen s conduc ed as pa o Wo k Package
6 ac i i ies can be ound in he deli e ables D6.1, D6.2, D6.3 and D6.4. Some o he modelling
de elopmen s a e s ill on-going: hese a e b ie ly desc ibed in he appendix, and will be
desc ibed in mo e de ail in he upcoming deli e able D6.5.
Fo all h ee supply chains, he scena io se con ains a baseline scena io (MS15) depic ing wha
hese supply chains migh look like i his o ical ends in demand, supply and ade a e p olonged
in he u u e. We hen conside al e na i e u u es, speci ic o each supply chains, and a ying
a ious aspec s ela ed o po en ial al e na i e sec o ial de elopmen s, pa ly in esponse o he
global clima e and biodi e si y c isis, bu also in es iga ing al e na i e assump ions abou land
managemen s a egies, echnological p og ess, demand p e e ences, and ade ne wo ks.
Al oge he , hese scena ios p o ide con as ed explo a i e u u es, p o iding a b oad pic u e o
wha u u e sec o ial ends migh look like, and wha he con ibu ion o key unde lying
assump ions a e. The li e a u e e iew unde pinning he p elimina y scena ios (p io o he
s akeholde wo kshops) a e p o ided in he INTRODUCTION sec ion, while he eedback om
he s akeholde s and he scena ios o each supply chains a e de ailed in he METHODS sec ion.
Fo each supply chain, da ase s o he model ou pu s a e p o ided as sepa a e Zenodo eco ds,
lis ed in he PROJECTS OUTPUTS ACHIEVED sec ion.
Scena io ou comes o soy supply chains
The baseline scena io depic s a business-as-usual u u e, based on he ‘Middle o he Road’
Sha ed Socioeconomic Pa hway (SSP2), complemen ed wi h key ecen land use policies in
B azil. Unde his business-as-usual scena io, he global demand o li es ock p oduc s and c op
7
p oduc s keeps g owing, and he expo s o soy-based p oduc s om B azil inc ease by 76% o e
he 2020-2050 pe iod, wi h inc eases o he EU bu p ima ily o Asia. While ood a ailabili y and
he alue o ag icul u al p oduc ion inc ease globally, u he biodi e si y losses om
ag icul u al ac i i ies a e p ojec ed. In B azil, al hough limi ed by he Amazon Soy Mo a o ium
and he Fo es Code, addi ional de o es a ion is p ojec ed, p ima ily h ough con e sion o
pas u es. The inc ease in soy p oduc ion is achie ed mainly h ough an expansion o e pas u es
and yield gains. Biodi e si y impac s om ag icul u e inc ease o bo h aqua ic and e es ial
ecosys ems, al hough a a slowe pace han ecen decades o e es ial biodi e si y. O e all,
his scena io poin s o a clea ade-o be ween economic and en i onmen al goals in he
coming decades, wi h conside able de elopmen oppo uni ies o he soy sec o in B azil and
a modes bu inc eased le el o impo s om soy p oduc s o he EU.
A i s al e na i e scena io explo es he impac o a global ood sys ems sus ainabili y
ansi ion owa ds bending global biodi e si y loss. Based on he Bending he Cu e s udy
(Leclè e e al., 2020), i en ails a comp ehensi e package o inc eased conse a ion and
es o a ion, demand-side measu es (lowe sha e o animal p oduc s in he die s, educed was e
and loss) and supply side measu es (sus ainably inc eased yields). I leads o simila ou comes
in e ms o ood a ailabili y globally, bu la ge dec eases in he alue o p oduc ion o li es ock
p oduc s, and o a mo e mode a e ex en , soy. The p oduc ion and ade o soy is p ojec ed
inc ease sligh ly abo e 2020 le els, and expo s o soy-based p oduc s om B azil inc ease by
abou 20% o e 2020-2050, while soya impo s om he EU dec ease o below 2020 le els. As a
esul , biodi e si y losses dec ease signi ican ly globally, as well as in B azil h ough a oided land
con e sion and he es o a ion o low in ensi y pas u es. O e all, his poin s o he isks e o s
owa ds he global en i onmen al c isis may pose o he soy and li es ock sec o , globally and
in pa icula o he soy sec o in B azil.
A second al e na i e scena io explo es he impac o an idealized ambi ious conse a ion
policy in B azil. I en ails a sensi i i y expe imen in which, on op o he baseline scena io no
con e sion o ei he o es o o he na u al land o ag icul u al land is allowed, o be e
unde s and he bounda y condi ions o he soy sec o in B azil. In con a y o he p e ious
scena io, i does no assume any o he global ood sys ems sus ainabili y ansi ion. This
scena io leads o e y simila ou comes han he baseline scena io, including a e y la ge
inc ease in soy p oduc ion and expo s om B azil (including some inc ease o he EU), as well
as alue o ag icul u al p oduc ion. Howe e , i leads a s ong dec ease in biodi e si y impac s,
e en lowe han in he p e ious scena io o land-use change-media ed impac s on e es ial
ecosys ems, o one ou o he wo ela ed indica o s. Al hough his scena io is p ojec ed o lead
o much lowe le els o pas u e es o a ion as compa ed o he p e ious scena io, his poin s o
a la ge ma gin o achie ing ambi ious biodi e si y ou comes in B azil h ough hal ing land
deg ada ion wi hou sac i icing po en ial economic oppo uni ies. This indica es some
po en ial mu ual in e es in such a scena io, be ween p oduce s in B azil willing o pu sue
economic g ow h, and EU consume s no willing o educe hei consump ion o animal
p oduc s.
A las al e na i e scena io explo es he long- e m impac s om he US-China ade dispu e on
he soy ma ke s. Based on ecen de elopmen s in soya expo s om USA o China, i explo es
he implica ions o an assumed cap o USA soya expo s o China o 25% below 2020 le els,
8
pic u ing long-las ing educed ma ke sha es o USA soya expo s o China h ough us and
eliabili y issues. In his scena io, we p ojec China o u n o B azil o compensa ing mos o
he impo sho all. This is p ojec ed o be achie ed wi h limi ed impac s on o he expo
des ina ion o B azil soy-based p oduc s, while he expo sho all om he USA is no
ealloca ed o o he des ina ions, in a con ex o e y compe i i e ma ke s. T ends in he impo
o soya-based p oduc s o he EU om di e en sou cing egions is expec ed o emain simila
o he baseline. Impac s on biodi e si y in B azil and elsewhe e, which a e signi ican ly
inc easing in he baseline, a e no p ojec ed o be signi ican ly highe in his scena io. Mode a e
losses in soy p oduce e enues a e p ojec ed in he USA as compa ed o he baseline, as a esul
o bo h p oduc ion and p ice declines. Unde he condi ion ha cu en key land use policies
o he soy sec o s ay in place, his poin s o he US-China ade dispu e being a much lowe
isk o B azilian ecosys ems han he g owing demand o li es ock p oduc s, in pa icula in
China.
Scena io ou comes o o es y supply chains
The baseline scena io depic s a business-as-usual u u e whe e bioene gy demand is ixed a
2020 le el, cons uc ion ma e ial demand is d i en by popula ion and GDP g ow h unde he
SSP2 socio-economic pa hway and plan a ion o es a ea is ixed a 2020 le el. An al e na i e
business-as-usual demand scena io, whe e plan a ion o es s can expand ou side he na u al
o es s a ea is also conside ed. Unde he business-as-usual scena ios, oundwood ha es
olumes a e p ojec ed o mode a ely inc ease by 2100. Ne expo s o wood-based p oduc s a e
ela i ely s able, wi h bo eal/ empe a e egions losing compe i i eness and opical egions
gaining some. A o es a ion up ake leads o an inc ease in o es ca bon s o age in he coming
decades, compensa ing o declining ca bon s o age in he exis ing na u al/semina u al o es
a ea. The global biodi e si y impac s om o es y diminish o e ime, and sligh ly educ ion is
obse ed mo e when allowing plan a ion o es expansion. This sugges s ha mode a e g ow h
in demand o wood-based p oduc s could be achie ed wi h an inc ease in o es ca bon
s o age and wi h less p essu es on biodi e si y.
A i s se o scena ios explo es he impac o an inc ease in demand o cons uc ion ma e ials
and/o bioene gy wi hou allowing o plan a ion o es s expansion. The global demand o
bioene gy is assumed o double be ween 2020 and 2100 as 90% o he new u ban popula ion is
assumed o li e in wooden buildings, wi h he ela ed demand inc ease p ima ily loca ed in
A ica and Asia. Highe demand o cons uc ion ma e ials leads o a 50% inc ease in ha es
olumes compa ed o he baseline, mos ly occu ing in Asia, A ica and La in Ame ica. Na u al
o es managemen is la gely in ensi ied, wi h up o 300Mha o na u al o es s being aken in o
p oduc ion, and nega i e implica ions o bo h o es ca bon balance and biodi e si y. Highe
bioene gy demand, on he o he hand, has a limi ed impac on ha es olumes as i is mainly
me h ough an expansion in ene gy c ops p oduc ion. Highe bioene gy demand esul s in an
inc ease in o es ca bon s o age compa ed o he baseline due o g owing ca bon s o age o
BECSS, and a educ ion in biodi e si y impac s o e ime, al hough a a lowe a e han in he
BAU scena ios. This sugges s ha s ong g ow h in demand o cons uc ion ma e ials could
exe signi ican p essu e on biodi e si y i me h ough in ensi ied na u al o es s
9
managemen whe eas highe demand o bioene gy has a limi ed impac as i mainly sou ced
om ene gy c ops expansion.
A second se o scena ios explo es he impac o a simila inc ease in demand o cons uc ion
ma e ials and/o bioene gy bu wi h he possibili y o expanding plan a ion o es s ou side
he na u al o es s a ea. This leads o an addi ional inc ease in ha es olumes as compa ed o
he baseline, due o lowe oundwood p ices as plan a ion o es s a e mo e e icien han
na u al/semina u al o es s. This also leads o an inc ease in he sha e o wood sou ced om
opical egions, whe e he p oduc i i y o o es plan a ions is highe . Impo an ly, plan a ion
expansion enables he elease o 200-300 million hec a es o semina u al o es s om
p oduc ion and hei es o a ion in o na u al o es s, bu leads o some decline in (biodi e si y
poo ) o he na u al land and ag icul u al land. Mo eo e , when conside ing a join inc ease in
cons uc ion ma e ials and bioene gy demand, a ailable plan a ion a ea and he e o e spa ed
na u al o es s a ea a e lowe (compa ed o he scena io wi h highe ma e ial demand only) as
ene gy c ops and plan a ion o es s compe e o he same land. Bu o e all, plan a ion o es
expansion leads o much be e ou comes o o es ca bon balance and biodi e si y han an
in ensi ica ion in na u al o es managemen (wi h impac s being close o baseline le els). This
highligh s he po en ial o a la ge plan a ion o es s expansion in suppo ing he de elopmen
o a wood-based bioeconomy while alle ia ing p essu e on na u al o es s and biodi e si y.
Howe e , his also poin s o a po en ial ade-o s be ween wood and bioene gy p oduc ion
g ow h due o land compe i ion be ween plan a ion o es s and ene gy c ops.
Scena io ou comes o c op aqua eed supply chains
The baseline scena io depic s a business-as-usual u u e, in which demand ends o blue ood
p oduc s ollow his o ical ends, he le el o wild ca ch emains cons an , and echnological
p og ess leads o a highe sha e o was e in ish meal and ish oil educ ion, a highe p oduc i i y
o ed aquacul u e sys ems, and an inc eased sha e o c ops in aqua eed equi emen s. As a
esul , he consump ion o blue ood p oduc s, and in pa icula , p oduc s sou ced om
eshwa e aquacul u e (e.g., ca ps) and un ed aquacul u e (e.g., c us aceans, ish and p awns),
inc ease, p ima ily h ough ed and un ed aquacul u e. While he demand o ish meal and ish
oil dec ease, he demand o c op aqua eed keep inc easing, bu a slowe pace han he
aquacul u e supply, and a he same pace han he o al demand o conce ned c ops. The sha e
o aqua eed use in he o al consump ion o ela ed c ops he e o e emains cons an o a ew
pe cen age poin s. Fu he loss o e es ial biodi e si y and an inc ease in nu ien losses om
eshwa e ed aquacul u e sys ems a e p ojec ed. This poin s o he g ow h o demand o blue
ood p oduc s, p ima ily supplied ia aquacul u e, as con ibu o o global g ow h in he
en i onmen al impac s om aising ood consump ion. This end is howe e mode a ed by
changes in aquacul u e eeding p ac ices, ha also edi ec p essu es om aqua eed demand
om ma ine and pelagic ish s ocks o e es ial ecosys ems.
A second se o 4 sensi i i y scena ios allows o explo e he ole o speci ic assump ions abou
he de elopmen o a ious o ms o aquacul u e and ela ed eeding p ac ices. In pa icula ,
po en ial inc ease in un ed aquacul u e a e equi ed o u u e le els o demand o speci ic
p oduc s (e.g., c us aceans) o be me , and i could also sligh ly alle ia e he need o ed
16
METHODS
Soy supply chains
O e all app oach
Goal: Fo soy supply chains, he main objec i e is o analyse he po en ial e olu ion o soy
ma ke s and o quan i y he po en ial impac o a ious go e nance mechanisms on soy
p oduc ion, ade and associa ed socio-economic and en i onmen al ou comes. In D7.2, we
ocus on analysing he impac o s ylized in e en ion scena ios cap u ing key sou ces o
unce ain ies o soy ma ke s a ound a business-as-usual (BAU) u u e. An addi ional se o
scena ios, explo ing he impac o speci ic policies in B azil and EU will be de eloped in D7.3.
Me hod summa y: The scena ios’ design elies on he S o y-and-simula ion app oach (Alcamo,
2001) wi h an i e a i e p ocess be ween expe -led s o yline de elopmen , quan i ica ion wi h
he GLOBIOM model and eedback om s akeholde s. The scena ios we e p ima ily de eloped
h ough a li e a u e e iew, as well as expe eedback on scena io ideas and key assump ions
(including p elimina y quan i ica ion), ga he ed du ing an online wo kshop.
Li e a u e e iew: As summa ized in he in oduc ion, he li e a u e e iew sugges s ha he
de elopmen o u u e global demand o animal p oduc s and ege able oils, and well as
sys em-wide in e en ions ac oss land use and ood sys ems owa ds global sus ainabili y goals
a e likely o inc easingly shape soy supply chains in he u u e. Recen ade ensions and a i
escala ion be ween he Uni ed S a es and China a e also expec ed o a ec soy ade and
en i onmen al ou comes in he coming decades.
Insigh s om he s akeholde wo kshop: A wo-hou online wo kshop ook place on 14 May
2025. I ga he ed 23 soy and ag icul u e expe s, om he p i a e sec o , public sec o ,
in e na ional o ganiza ions and NGOs om bo h B azil and Eu ope. The wo kshop included: 1)
a 20min p esen a ion om IIASA on model imp o emen s, po en ial scena io op ions and
p elimina y esul s, ollowed by a Q&A; 2) a i s 20min b eakou session whe e expe s we e
asked o sha e hei ision o he soy sec o by 2050 and a epo ing back in plena y; 3) a second
20min b eakou session whe e expe s we e asked o iden i y hei p e e ed scena io op ions
and discuss he likely impac o he selec ed policies/s ylized in e en ions on soy expo s and
biodi e si y, ollowed by a epo ing back in plena y. Expe s men ioned popula ion dynamics,
he e olu ion in China’s soy demand, die a y shi s away om mea owa ds plan -based
p oduc s, de o es a ion egula ions and he Uni ed S a es-China ade wa as main ac o s likely
o impac he soy sec o in he coming decades. Se e al o hese elemen s a e conside ed in he
s ylized scena ios de eloped in D7.2. S akeholde s also sha ed eedback on scena ios
conside ing speci ic policy in e en ions in B azil and EU, indi idually and in a ious
combina ions (e.g., ASM, Fo es Code, EUDR, and EU-Me cosu ) ha will be used o in o m D7.3.
Scena io desc ip ion
Based on insigh s om he li e a u e and he s akeholde wo kshop, we explo e he socio-
economic and en i onmen al impac s o a BAU u u e and o h ee s ylised scena ios cap u ing
17
key sou ces o unce ain ies o soy ma ke s. The main assump ions behind hese scena ios a e
summa ized in Table 1 while a mo e de ailed desc ip ion is p o ided below.
Table 1 - Summa y o assump ions o soy supply chain scena ios
Scena io name
Assump ions
Ques ions explo ed
SSP2
P olonga ion o his o ical ends
(no policy change)
Wha does a BAU u u e mean o
soy ma ke s and biodi e si y?
IAP
Conse a ion and es o a ion,
supply and demand-
side e o s
aligned wi h he KMGBF goal o
e e sing global biodi e si y loss
om land use change by 2050
Wha does eaching ambi ious
biodi e si y goals means o soy
ma ke s?
T .Dis
China’s impo s o US soy-based
p oduc s capped a 75% o 2020
alue
Wha could be he long- e m
consequences o he US-China ade
wa on soy ma ke s and biodi e si y?
ZNL_B a
Ze o absolu e con e sion o o es
and o he na u al lands o
ag icul u al land use in B azil a e
2020
Wha s ong land conse a ion in
B azil means o soy p oduc ion and
ade?
As illus a ed in Table 1, ou scena ios a e conside ed, namely:
• BAU scena io (SSP2): The ‘Middle o he Road’ Sha ed Socioeconomic Pa hway (SSP) 2
is used as a baseline scena io. I pic u es a p olonga ion o his o ical ends in
popula ion, die a y p e e ences, ade and ag icul u al p oduc i i y in o he u u e.
Mo e speci ically, SSP2 ep esen s a wo ld in which he human popula ion peaks a 9.4
billion by 2070, economic g ow h is mode a e and une en, and globaliza ion con inues
wi h slow socioeconomic con e gence be ween coun ies. C op yields inc ease globally
wi h a pa ial con e gence be ween high- and low-yield egions, while no addi ional
e o a e made owa ds mi iga ing he iple plane a y c isis (i.e., clima e change,
biodi e si y loss, pollu ion) beyond a educ ion in de o es a ion om o he sou ces han
ag icul u al land use expansion (Popp e al., 2017).
• In eg a ed ac ion po olio (IAP) scena io: his scena io conside s he global
implemen a ion o ood and land use sys ems-wide in e en ions which enables o
e e se he global e es ial biodi e si y ends caused by habi a con e sion by 2050
while eeding he g owing human popula ion (Leclè e e al., 2020). The in eg a ed ac ion
po olio includes a mix o supply-side (sus ainable inc ease in c op yields, ollowing he
‘G een oad’ SSP1 scena io ins ead o SSP2), demand-side (die shi owa ds lowe
sha e o animal calo ies han in he baseline in egions o high consump ion1, and was e
1 I is assumed ha he consump ion o animal p oduc s is dec eased by hal as compa ed o he le els assumed in
he baseline o 2050, in all egions bu he Middle-Eas and Sub-Saha an A ica (pa o he AME agg ega ed egion),
India and Paci ic islands (pa o SAS agg ega ed egion), as well as Sou heas Asia (SEA agg ega ed egion), and
eplaced by an inc eased consump ion o plan -based p oduc s o each he same le el o consump ion in calo i ic
e ms.
18
educ ion om a m o o k) and land conse a ion and es o a ion measu es (inc eased
ex en o p o ec ed a eas o abou 40% o e es ial a eas by 2030, inc eased
e ec i eness o p o ec ed a eas p e en ing any biodi e si y-de imen al addi ional land
use con e sion wi hin hose a e 2020, p og essi e es o a ion eaching abou 450
million hec a es by 2050, and pe asi e conside a ion o biodi e si y in land use
planning).
• T ade dis up ion (T .Dis.) scena io: his scena io caps China’s impo s o soy-based
p oduc s om he Uni ed S a es a 75% o 2020 alue om 2030 onwa ds. I ep esen s
a plausible assump ion abou he expec ed long- e m e ec s o he ecen US-China
ade dispu e, which is expec ed o e ode he compe i i eness o US soy expo s o
China. This ep esen s mo e han a 25% decline in US soy expo s o China a e 2020,
compa ed o he BAU scena io. This could be conside ed as qui e pessimis ic, gi en ha
China’s soy impo demand is p ojec ed o con inue inc easing in he coming decades.
• Ze o na u al land loss (ZNL_B a) scena io: his scena io assumes ze o absolu e
con e sion om o es and non- o es na u al land o ag icul u al land in B azil a e
2020. I is used o assess he p oduc ion and expo expansion po en ial o B azil’s soy
sec o i e y ambi ious land conse a ion ambi ions we e me . I is no mean o
ep esen a ealis ic policy de elopmen bu is a he in ended o cap u e he bounda y
condi ions o B azil’s soy sec o .
These ou scena ios we e discussed du ing he s akeholde wo kshop and me wi h in e es , as
hey we e conside ed o cap u e well he key sou ces o unce ain ies o he u u e e olu ion
o soy ma ke s. Indeed, soy expe s conside ed he u u e e olu ion in China’s soy demand,
die a y shi s away om mea owa ds plan -based p oduc s, de o es a ion egula ions and he
Uni ed S a es-China ade wa as key ac o s likely o impac he soy sec o in he coming
decades.
As pa o an e o o imp o e he ep esen a ion o land use dynamics in B azil, key policy and
p i a e sec o in e en ions including he Fo es Code and he Amazon Soy Mo a o ium we e
pa ame e ized in he model. As u he de ailed in appendix, he pa ame e iza ion o he Fo es
Code assumes es ic ions on illegal de o es a ion in he Amazonia and Ce ado biomes om
2020 onwa ds, adjus ed by en o cemen p obabili ies, oge he wi h es o a ion e o s om
2030 onwa ds modelled as a con e sion o c opland and pas u e in o p o ec ed o es . Fo he
Amazon Soy Mo a o ium no expansion in he land a ea dedica ed o soy in he Amazon biome
is assumed. These ea u es apply bo h o he BAU scena io and he h ee explo a i e scena ios.
Quan i ica ion
The GLOBIOM economic model is used o gene a e p ojec ions o ma ke de elopmen
indica o s and en i onmen al and socio-economic indica o s om he ini ial yea (2000) un il
2050 (wi h a 10-yea ime s ep) o he baseline and he explo a i e scena ios. Resul s om 2020
19
onwa ds o main global egions, and a subna ional scale o B azil, a e analysed in he esul s
sec ion o his deli e able.
Ou comes unde he baseline and he s ylised policy scena ios a e analysed o di e en
indica o s o ood consump ion, p oduc ion and ade and se e al en i onmen al indica o s
including land use, wa e use, GHG emissions and biodi e si y loss. Biodi e si y loss indica o s
a e based on li e cycle assessmen (LCA) me hods and ha e been de eloped in he con ex o
D6.3. These indica o s ha e been linked o he GLOBIOM model, and hei alue has al eady
been p ojec ed o a BAU SSP2 scena io in D6.4. The analysis o he BAU and he policy scena ios
is ca ied ou a he global and egional le els ( o i e agg ega ed egions and he EU)2 as well
as a he coun y and biome le el o B azil.
Fo soy in B azil, in addi ion o he biodi e si y impac o a ming quan i ied wi h GLOBIOM, he
model was expanded o also include ups eam and downs eam supply chain impac s based on
he esul s o LCA (see D6.4). These include biodi e si y impac s associa ed wi h inpu
p oduc ion (e.g., e ilize ), soy p ocessing and soy anspo a ion, bo h domes ically and
ab oad. This enables o ge a mo e comple e pic u e o he ull ange o impac associa ed wi h
soy supply chains in B azil.
Fo es supply chains
O e all app oach
Goal: Fo o es -based supply chains, he main goal is o quan i y how he po en ial u u e
de elopmen s o wood demands will impac on oundwood ha es olumes, o es
managemen a eas, land use changes, o es ca bon balance and biodi e si y. This includes also
unde s anding how he o es sec o compe i i eness will de elop ac oss di e en egions and
how his would a ec global sus ainabili y. In his deli e able, we ocus on explo a i e scena ios
pic u ing plausible u u e de elopmen s in he sec o ial demands and land uses impac s, as a
complemen o mo e p elimina y scena ios p esen ed in D6.2.
Me hods summa y: To design and quan i y he scena ios, we ely on a se ies o scena ios
na a i es, designed by sec o ial expe s and quan i ied wi h he GLOBIOM model, including he
eedback om s akeholde s. The scena ios we e i s ly de eloped h ough li e a u e e iew and
sec o ial expe s, a e wa ds hey we e u he e ined by in eg a ing s akeholde s’ inpu on
scena io ideas and key assump ions (including p elimina y quan i ica ion), ga he ed du ing an
online wo kshop. The GLOBIOM model was u he imp o ed o accommoda ing he scena ios
speci ic needs ela ed o ca bon dynamics and biodi e si y loss. Model imp o emen s, including
he addi ion o o es age-class dynamics, ha es ed wood p oduc s (HWP) and Bioene gy wi h
Ca bon Cap u e and S o age (BECCS) ca bon accoun ings, a e desc ibed in annex o his epo .
2 The i e agg ega ed egions conside ed a e No he n Ame ica (NAM), O he Sou h Ame ica (OSA) (excl. B azil),
A ica and Middle Eas (AME), Sou h Asia (SAS – including China), and he Res o he Wo ld (ROW).
20
Insigh s om he li e a u e e iew: A li e a u e e iew was used o iden i ying u u e majo
de elopmen s ha will shape he u u e bioeconomy in e ms o wood demands and ca bon
s o age po en ials. The mos in e es ing eme ging de elopmen s include he g ow h o wood
demands in he building sec o , disclosing he po en ial o s o ing ca bon in long li ed ha es ed
wood p oduc s (HWP). Acco dingly, we de eloped al e na i e scena ios o wood use in
buildings, one mo e conse a i e whe e imbe demands a e aligned o GDP/POP g ow h, and
one whe e we conside ed also u u e “ imbe ci ies” de elopmen and aligned o explo a o y
scena ios in (Mish a e al., 2022), d i en by u baniza ion in di e en global egions. Addi ionally,
we conside ed u u e al e na i e g ow hs in demand o BEECS associa ed wi h bioene gy
scena ios (IIASA, 2021), gi en ha bioene gy demand is ano he ele an d i e o wood-based
mi iga ion.
Insigh s om he s akeholde wo kshop: A 90-minu e online wo kshop was conduc ed on
Decembe 5 h 2024, in ol ing 10 s akeholde s in oughly equal p opo ions om NGOs,
uni e si ies and indus y & p i a e sec o , p edominan ly based Eu ope bu wi h pa icipan s
also based in La in Ame ica. The wo kshop included an in oduc ion session on he scope o he
wo k, a session dedica ed o ga he ing s akeholde iews on he ision o he sec o by 2050,
and a session dedica ed o collec ing eedback on he main scena io dimensions, ela ed
assump ions and quan i ied ou comes. Mo e de ails on he wo kshop o ganiza ion can be ound
in appendix. The s akeholde s ound he ini ial scena io scope ele an bu also exp essed
in e es in conside ing mo e “mul i pu pose” managemen s in o es s ecosys ems oge he wi h
scena ios ha could add ess p o ec ion o p ima y o es s and es o a ion ambi ions in na u al
ecosys ems. Mul i unc ional o es s a e included in he managemen op ions conside ed in he
modelling o ou scena ios. Fo u he expanding he conse a ion and mul ipu pose alue o
na u al/semina u al o es s, we ha e e ined ou p elimina y scena ios (D6.2), by ocusing on
o es plan a ions expansion ou side o es a ea and es ablishing new plan a ions only on
abandoned ag icul u al land (i.e. excluding na u al land). This app oach is able o incen i izing a
win-win solu ions able o educing p essu e on p ima y and semina u al o es s wi hou
deple ing o he aluable ecosys ems. A he same ime, we ha e conside ed he expansion o
a o es a ion o ca bon and biodi e si y as a es o a ion op ion. These di e en assump ions
oge he go in he di ec ion o ca bon-o ien ed o es managemen s and bio-based solu ions as
sugges ed by s akeholde s. In e es in mo e socioeconomic indica o s was also aised by
s akeholde s, while ou modelling app oach allows o disclose some mac oeconomic ends, like
he ones ela ed o o es sec o compe i i eness ac oss di e en global egions, including wood
p oduc s p oduc ion and ade, his app oach does no allow o co e in mo e de ailed indica o s
ela ed o local wel a e (i.e. local employmen , wages, heal h), howe e , he o e all egional
sec o ial de elopmen can be used as an indica o o he o es sec o economic g ow h.
Sugges ions included also o conside na i e species and local communi ies; howe e , his
aspec is ha dly cap u ed wi hou a subs an ial e inemen o he modelling amewo k in
speci ic egions and would go beyond he scope o he scena ios p esen ed in his deli e able.
21
Scena io desc ip ion
We conside a o al o 8 scena ios (Table 2), which di e in e ms o he assump ions made
ega ding o es managemen and wood demand. All scena ios a e consis en wi h he RCP1p9
clima e mi iga ion a ge and he SSP2 socio-economic de elopmen pa hway. The BASE
scena io is a “business-as-usual” scena io whe e o es managemen is based on cu en
managemen p ac ices and he e is no de elopmen o ene gy use and limi ed de elopmen o
ma e ial use.
Table 2. Summa y o assump ion o o es supply chains scena ios
Scena io name
S o yline
Bioene gy
Cons uc ion
ma e ials
Fo es
plan a ions
BASE
“Business as usual”
Fixed 2020
le el
SSP2 demand
Fixed 2020 le el
BIO
High bioene gy wi h
cu en managemen
RCP1p9 ne
ze o
SSP2
demand
Fixed 2020 le el
CON
High cons uc ion wi h
cu en managemen
Fixed 2020
le el
Timbe ci ies
Fixed 2020 le el
BIO&CON
High bioene gy and
cons uc ion wi h
cu en managemen
RCP1p9 ne
ze o
Timbe ci ies
Fixed 2020 le el
PLA
Business as usual wi h
o es plan a ions
Fixed 2020
le el
SSP2
demand
Expansion
ou side cu en
o es a ea
PLA BIO
High bioene gy wi h
o es plan a ions
RCP1p9 ne
ze o
SSP2
demand
Expansion
ou side cu en
o es a ea
PLA CON
High cons uc ion wi h
o es plan a ions
Fixed 2020
le el
Timbe ci ies
Expansion
ou side cu en
o es a ea
PLA BIO&CON
High bioene gy and
cons uc ion wi h
o es plan a ions
RCP1p9 ne
ze o
Timbe ci ies
Expansion
ou side cu en
o es a ea
The scena ios a e based on combina ions o h ee dimensions, including he bioene gy demand
(2 a ia ions = ixed 2020 le el, RCP1p9 ne ze o), cons uc ion ma e ial demand (2 a ia ions =
SSP2 demand, Timbe ci ies) and o es plan a ion a ea expansion (2 a ia ions = ixed 2020
le el, Expansion ou side cu en o es a ea), as desc ibed in he ollowing sec ions.
Unde he BASE scena io, we assume a business a usual de elopmen which conside s
bioene gy demand ixed o 2020 le el, cons uc ion ma e ials demand ollowing SSP2 demand
22
de elopmen (i.e. “middle o he oad” socioeconomic g ow h scena io) and o es plan a ion
a ea is ixed a 2020 le el.
Bioene gy demand
S anda d high mi iga ion bioene gy demand in GLOBIOM is based on he RCP1p9 o e shoo
scena io, whe e global bioene gy demand inc ease om 57 exajoules (EJ) in 2020 o 226 EJ in
2100 (IIASA 2018). This scena io was calcula ed abou 10 yea s ago unde mino sus ainabili y
cons ain s on he land-use sec o . Nowadays he gene al iew o he in eg a ed assessmen
model (IAM) communi y is ha his le el o bioene gy demand canno be eached i s ic e
land-use sus ainabili y cons ain s a e applied (Wu e al., 2019). The e o e, we apply a lowe
le el bioene gy demand based on he RCP1p9 ne -ze o scena io.
Speci ically, wo scena ios a e conside ed o bioene gy demand (Figu e 1). In he i s scena io
(RCP1p9 ix2020) bioene gy demand is ixed a 2020 le el 57 EJ. This scena io implici ly assumes
ha addi ional enewable ene gy needed o he RCP1p9 mi iga ion a ge comes om o he
enewable ene gy sou ces such as sola and wind powe ins ead o bioene gy. The second
scena io (RCP1p9 ne ze o) is he MESSAGE RCP1p9 ne -ze o scena io (EN_NPi2050_500) whe e
global bioene gy demand inc ease om 57 EJ in 2020 o 116 EJ in 2100 (IIASA, 2021).
Figu e 1. Global bioene gy demand scena ios
No e: 1 EJ p ima y ene gy ≈140 Mm3 woody biomass
In 2020, he majo i y o bioene gy demand is loca ed in Asia and A ica whe e bioene gy
p o ides he majo ene gy sou ce o hea ing and cooking o households (Figu e 2). In he ne
ze o scena io, bioene gy demand inc eases mos in Asia and La in Ame ica which ha e he
la ges biomass po en ials o bioene gy p oduc ion.
0
50
100
150
200
250
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
EJ/y
Global bioene gy demand
RCP1p9 ix 2020 RCP1p9 ne ze o RCP1p9 o e shoo (no eachable)
23
Figu e 2. Global bioene gy demand di ided by egion
Wood-based cons uc ion ma e ials demand
Two demand scena ios a e conside ed o wood-based cons uc ion ma e ials (Figu e 3). In he
i s scena io, he demand o wood-based cons uc ion ma e ials is based on he demand o
semi inished p oduc s, which is d i en by popula ion and GDP g ow h om he SSP2 scena io.
The second scena io is he “Timbe Ci ies” scena io om (Mish a e al., 2022), whe e 90% o he
new u ban popula ion is assumed o li e in wooden buildings. The numbe o new u ban
buildings is based on he amoun o people mo ing om he coun yside o ci ies and is
app oxima ed by he egional u baniza ion a es om o he SSP2 scena io.
Figu e 3. Global wood-based cons uc ion ma e ials demand scena ios
No e: 1 M on cons uc ion ma e ials ≈2 Mm3 woody biomass
0
20
40
60
80
100
120
EJ/y
RCP1p9 ix 2020
EU27 ASIA
A ica Fo me So ie Union
La inAme icaRes No hAme ica
B azil
0
20
40
60
80
100
120
EJ/y
RCP1p9 ne ze o
EU27 ASIA
A ica Fo me So ie Union
La inAme icaRes No hAme ica
B azil
0
200
400
600
800
1000
1200
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
M on/y
Wood-based cons uc ion ma e ials
SSP2 baseline Timbe ci ies (Mish a 2022)
24
As showed in Figu e 4, in he “Timbe Ci ies” scena io mos new wooden buildings a e buil in
A ica and Asia a he han in egions which adi ionally ha e wooden buildings such as he
Eu opean Union (EU) and No h Ame ica. This is because u baniza ion a es a e much highe in
Asia and A ica han in he EU and No h-Ame ica.
Figu e 4. Wood-based cons uc ion ma e ials demand by global egions
Plan a ion o es s and ene gy c ops plan a ions
Two scena ios a e conside ed o plan a ion o es s (Figu e 5). The di e ence o he scena ios
p esen ed in D6.2 is ha na u al o es s a e no allowed o be con e ed in plan a ions and
plan a ions expansion is limi ed o o he na u al land. In he i s scena io, plan a ion o es s
a ea is ixed a he 2020 le el (132 Mha). In he second scena io, plan a ion o es s a ea can be
expanded unde he speci ied land-use change cons ain s, which allows o inc ease plan a ion
o es s a ea up o 200-300 Mha in 2100 depending on he demand scena io. New plan a ions
a e mos ly es ablished in he opical egions, because biomass yields a e highe and land-use
compe i ion is lowe in opical zone han in bo eal and empe a e zones.
0
200
400
600
800
1000
1200
M on/y
SSP2 BASE
EU27 ASIA
A ica
Fo me So ie Uni
La inAme icaRes No hAme ica
B azil
0
200
400
600
800
1000
1200
M on/y
Timbe Ci ies
EU27 ASIA
A ica Fo me So ie Union
La inAme icaRes No hAme ica
B azil
25
Figu e 5. Global plan a ion o es s a ea acco ding o di e en plan a ions scena ios
Besides plan a ion o es s, GLOBIOM also includes ene gy c ops plan a ions, which compe e
wi h plan a ion o es s o he same land a eas (Figu e 6). The di e ence be ween hem is ha
plan a ions o es s p oduce oundwood and a e classi ied as o es a ea, while ene gy c ops
plan a ion p oduce bioene gy c ops and a e classi ied as ag icul u al land a ea
Figu e 6. Global ene gy c ops plan a ion a ea de elopmen acco ding o di e en bioene gy demand scena ios
Quan i ica ion
The scena ios a e quan i ied wi h a e ined e sion o he GLOBIOM economic model ha
includes an enhance ep esen a ion o he o es sec o and i is used o gene a e p ojec ions
o ele an ou comes un il 2100 unde he di e en scena ios. Mo e de ails on he modelling
amewo k can be ound in he deli e able D6.2 and in he appendix. Addi ional model
0
50
100
150
200
250
300
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Mha
Plan a ion o es s
Fix 2020
F ee expansion (low cons uc demand)
F ee expansion (high cons uc demand)
0
100
200
300
400
500
600
700
800
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Mha
Ene gy c ops plan a ions
RCP1p9 ix 2020 RCP1p9 ne ze o RCP1p9 o e shoo (no eachable)
32
Figu e 7. P ojec ed ends in ood a ailabili y pe capi a (in ene ge ic con en , kcal/cap/day) o i e agg ega ed
and wo zoomed egions (B azil BRA and Eu opean Union EUE). The colo s di e en ia e animal-based ( u quoise)
s plan -based ood p oduc s (yellow).
P oduc ion and ne ade
Figu e 8 displays ends in p oduc ion and ne ade by agg ega ed egions o li es ock
p oduc s, soy and o he c ops.
In he BAU SSP2 scena io, ag icul u al p oduc ion is p ojec ed o inc ease globally be ween 2020
and 2050, wi h mos li es ock and c op ou pu g ow h occu ing in SAS, AME and OSA. Soy
p oduc ion is p ojec ed o g ow in all p oducing egions (i.e., NAM, BRA and OSA, SAS), bu mo e
s ongly in B azil - he wo ld’s la ges p oduce - which is expec ed o accoun o o e hal o
global ou pu g ow h.
Ne expo e s o li es ock p oduc s a e expec ed o inc ease hei ade su pluses be ween 2020
and 2050 while o o he c ops, ne expo s a e p ojec ed o inc ease in B azil bu decline in
NAM and OSA. As displayed in Figu e 8 o ne expo s o soya bean only, and in Figu e 9 o ne
bila e al ade in all soy-based p oduc s (in soybean equi alen ) be ween agg ega ed egions,
ne expo e s o soy and soya-based p oduc s a e p ojec ed o inc ease hei ade su pluses,
wi h B azil consolida ing i s posi ion as he wo ld’s la ges expo e and inc easing i s ne expo s
o soya-based p oduc s om 97 M in 2020 o o e 171 M in 2050 (+76%). EUE main ains a
su plus o li es ock p oduc s, a ela i ely neu al ade posi ion o o he c ops han soy, and
an inc eased de ici o soy- ela ed p oduc s (ne impo s inc ease om 25 M o soya-
equi alen in 2020 o 33 M o soya-equi alen by 2050). AME, in u n, is expec ed o see i s
33
ag icul u al ade de ici widen in he coming decades whe eas SAS is p ojec ed o educe i s ne
impo s o bo h li es ock and c op p oduc s (excl. soy-based p oduc s) due o s ong p oduc ion
g ow h. In con as , SAS impo s o soya-based p oduc s ep esen he majo i y o global
inc eases in ne impo s o soya-based p oduc s, om 120 M o soya equi alen s o abou 200
M o soya equi alen s in 2050.
In he IAP scena io, li es ock p oduc ion d ops in all egions compa ed o he 2050 BAU scena io,
wi h s ong decline in NAM, EUE, BRA and OSA, and ROW (all a ound 40%). Global soy
p oduc ion - which is mos ly used as animal eed - also dec ease compa ed o he baseline (wi h
decline in BRA and OSA, and SAS bu some inc ease in NAM, whe e soy la gely con ibu es o
he inc ease in ege able oil consump ion equi ed o compensa e o losses in animal calo ies,
u he incen i ized by a dec ease in ege able oil p ice – see Figu e 10). I should be no ed ha
he dec ease in soy p oduc ion is lowe han p ojec ed wi h GLOBIOM in (Leclè e e al., 2020),
due o he imp o ed ep esen a ion o oilseed supply chains in CLEVER. As soybean meal (used
o li es ock eed) and soy oil (used o o he pu poses) a e join ly p oduced h ough c ushing,
educ ions in eed demand ollowing he assumed die a y shi ansla e in o smalle educ ions
in soy p oduc ion. In con as o he la ge inc ease p ojec ed be ween 2020 and 2050 in he BAU
scena io (+76%), o al ne expo s o soy-based p oduc s om B azil o o he agg ega ed egions
inc ease by only 20% o e he same pe iod in he IAP scena io, and e en dec ease o expo s
o he EU (- 34%).
In he IAP scena io, he p oduc ion o o he c ops also declines sligh ly om 2050 baseline le els
- al hough a a lowe a e han soy p oduc ion - as he educ ion in demand o eed c ops (and
associa ed educ ion in p oduc ion) o se s he inc ease in demand o plan -based ood. In line
wi h he obse ed ends in p oduc ion, ne ade in li es ock and c op p oduc s (incl. soy) is
below he 2050 BAU (in all egions).
In he T ade dis up ion scena io, soy p oduc ion in NAM declines om 2050 baseline le el (- 15
M ), wi h a co esponding inc ease in p oduc ion in B azil (+ 11 M ) and OSA (+2.6 M ).
Acco dingly, ne soy expo s om NAM dec ease (due o he decline in expo s o SAS, sligh ly
compensa ed by inc eases in expo s o AME, EUE and OSA) and hose om B azil and OSA
inc ease. This is in line wi h wha has been obse ed du ing he ecen ade wa s be ween he
Uni ed S a es and China, whe e B azil abso bs China’s soy demand ha is no me by he Uni ed
S a es by inc easing i s soy p oduc ion and expo s, and limi ed e-alloca ion o USA expo s o
o he ma ke s. This con i ms he po en ial isks his ade dispu e could ha e o biodi e si y-
ich ecosys ems in B azil, and soya p oduce s in he USA. Howe e , i should be no ed ha he
p ojec ed inc ease in soy expo s om B azil o China in he BAU is al eady e y high, he eby
highligh ing ha inc eased demand o soy in China cons i u es a la ge sou ce o isk o
B azilian ecosys ems.
Ze o con e sion o o es and o he na u al lands in B azil only has a small impac on ag icul u al
p oduc ion in he coun y; wi h soy and li es ock ou pu sligh ly declining om 2050 baseline
le els (bu inc easing om 2020 le els). Acco ding, B azil ne expo s o soy and li es ock
p oduc s only sligh ly lowe inc eases han o e 2020-2050 han in he baseline, by 4.5 M and
162 housand ons, espec i ely. This bene i s o he soy p oducing egions such as NAM and
OSA, whose ne expo s mode a ely inc ease as compa ed o he 2050 BAU.
34
Figu e 8. P ojec ed ends in p oduc ion and ne ade (physical olume, in million ons o esh ma e ) o i e
agg ega ed egions and wo zoomed egions (B azil BRA and Eu opean Union EUE). The colou s di e en ia e
li es ock p oduc s (LSP, in u quoise) om soy (SOY, in pu ple) and o he p ima y c ops (OCRP, in yellow).
Figu e 9. P ojec ed ends in ne ade o soy-based p oduc s (physical olume, in million ons o p ima y
p oduc s equi alen ) o i e agg ega ed impo ing egions and six expo ing egions (incl. B azil).
35
Subs i u ions in oilseed ma ke s
Figu e 10 shows ends in he consump ion o soy oil, o he ege able oils, soy p o ein meals
and o he p o ein meals. I enables us o see po en ial subs i u ion e ec s be ween p oduc s.
Vege able oils: Soy oil is he main cooking oil consumed in NAM and BRA while o he ege able
oils domina e in o he egions (e.g., palm oil in SAS and AME, sun lowe oil, and apeseed oil in
EUE). In a BAU u u e, global ege able oils consump ion is p ojec ed o inc ease be ween 2020
and 2050 – wi h s ong g ow h expec ed in SAS and AME due o popula ion g ow h and inc ease
in pe capi a consump ion. In mos egions, he consump ion o soy oil inc eases a a highe a e
han ha o o he ege able oils, wi h an inc ease o e ime in i s sha e in o al consump ion
(incl. in SAS, BRA and OSA, EUE, and ROW). This sugges s ha gains in soy compe i i eness in
ege able oil ma ke s will be a d i e o u u e g ow h in soy p oduc ion, pa ly d i en by he
demand o soy p o ein meal o animal eed.
In he IAP scena io, global ege able oil consump ion inc eases sligh ly om he 2050 BAU o
compensa e o he decline in li es ock calo ies, and as a esul o an o e all dec ease in c op
commodi ies p ices. Howe e , he sha e o soy oil in o al ege able oils consump ion declines
compa ed o he 2050 BAU, while he consump ion o o he ege able oils inc eases. This
subs i u ion is due o he assumed shi in die away om animal p oduc s, which leads o a
educ ion in demand o all eed p oduc s, including soybean meal.
The ZNL and he ade dis up ion scena ios a e p ojec ed o only ma ginally di e om he 2050
BAU in e ms o he subs i u ion be ween ege able oils.
P o ein meals: Soybean meals accoun o abou h ee qua e o global p o ein meal
consump ion. In he BAU scena io, he consump ion o p o ein meals is p ojec ed o inc ease
globally be ween 2020 and 2050, and mos no ably in SAS, OSA and AME due o an inc ease in
li es ock p oduc ion and eed in ensi ica ion in de eloping coun ies. Abou 80% o he global
inc ease in p o ein meal consump ion is expec ed o come om soybean meals.
In he IAP scena io, p o ein meal consump ion declines in all egions compa ed o he 2050 BAU
(excep in NAM) due o a shi in die s away om li es ock p oduc s and associa ed decline in
li es ock p oduc ion (see Figu e 8). Mos o he global decline in p o ein meals consump ion
comes om soybean meals while he consump ion o o he p o ein meals inc eases in some
egions (e.g., SAS, OSA, and ROW). In NAM, howe e , he inc ease in soy oil consump ion leads
o an inc ease in he consump ion o soybean meal (al hough epo ed as consump ion in Figu e
10, hal o i emains unused due o he d op in domes ic and global eed demand).
The ZNL and he ade dis up ion scena ios a e p ojec ed o only ma ginally di e om he 2050
BAU in e ms o he subs i u ion be ween p o ein meals.
36
Figu e 10. P ojec ed ends in consump ion (physical olume, in million ons o esh ma e ) o i e agg ega ed
egions and wo zoomed egions (B azil BRA and Eu opean Union EUE). The colo s di e en ia e soy oil (SoyO, in
u quoise), om o he ege able oils (OVegO, in pu ple) and soy p o ein meals (SoyP oM, in yellow) om o he
p o ein meals (OP oM, in ed).
Value o p oduc ion
The p ojec ed alue o p oduc ion (USD 2000) o p ima y ag icul u al p oduc s (i.e., excluding
seconda y p oduc s such as ege able oils and p o ein meals) is displayed in Figu e 11. I is
calcula ed as p oduc ion olume mul iplied by p oduce p ices, and p o ides an unde s anding
o changes in he economic alue o ag icul u al p oduc ion. In mos egions, li es ock p oduc s
ep esen a la ge sha e in he alue o p oduc ion han in physical olume, due o ela i ely
highe ma ke p ices.
In he BAU SSP2 scena io, he o al alue o ag icul u al p oduc ion is expec ed o inc ease in all
egions be ween 2020 and 2050, wi h s ong inc ease in he p oduc ion alue o li es ock and
c ops in AME and SAS. The alue o soy p oduc ion is p ojec ed o decline in NAM (due o a
p ojec ed p ice dec ease) and o inc ease in BRA and OSA.
In he IAP scena io, he alue o p oduc ion d ops in all egions compa ed o he 2050 BAU
mainly d i en by a s ong decline in he alue o li es ock p oduc ion. Li es ock p oduc ion alue
dec eases a a highe a e han physical olume (a a ound -60% in EUE, NAM, ROW, BRA and
OSA) due o a d op in ma ke p ices. The alue o soy p oduc ion also declines in BRA and OSA
(a a simila a e and sligh ly highe a e han physical olume, espec i ely) as well as in NAM,
despi e an inc ease in p oduc ion in his egion.
37
In he T ade dis up ion scena io, he alue o soy p oduc ion declines a a highe a e han
physical olume in NAM (-15% s -9%) as soy p ices d op due o a educ ion in Chinese impo
demand. This illus a es he ulne abili y o US soy p oduce s o ade dispu e wi h China. Soy
p oduc ion alue, in u n, inc eases in BRA (+6%) and OSA (+4%) – b oadly in line wi h changes
in physical olume, indica ing limi ed p ice e ec s on global ma ke s.
In he ZNL scena io, ag icul u al p oduc ion alue in B azil sligh ly inc ease om he 2050 BAU
due o an inc ease in he alue o li es ock p oduc ion as p ice inc eases mo e han p oduc ion
declines. This highligh s ha , in addi ion o he impac s on p oduc ion le els, p ice eac ions a e
impo an o conside when es ima ing he oppo uni y cos (in e ms o sec o ial economic
ou pu ) o ambi ious conse a ion e o s in B azil. The o e all impac in e ms o economic
ou pu migh e en be posi i e a he sec o ial le el, wi h limi ed changes in he dis ibu ion o
economic oppo uni ies ac oss ag icul u al sub-sec o s.
Figu e 11. P ojec ed ends in alue o p oduc ion (in billions o USD2000) o i e agg ega ed egions and wo
zoomed egions (B azil BRA and Eu opean Union EUE). The colo s di e en ia e li es ock p oduc s ( u quoise, LSP)
om soy (pu ple, SOY) and o he p ima y c ops (yellow, OCROP).
Land use
As shown in Figu e 12, he p ojec ed changes in land use om 2020 o 2050 a y ac oss egions,
wi h ela i ely s able land use in NAM, EUE and ROW and u he con e sion o o es and o he
land o c opland and pas u e in o he egions (BRA and OSA, AME, SAS). This is an expec ed
pa e n o SSP2 (e.g., (F icko e al., 2017; S eh es e al., 2019)), wi h p opo ionally lowe
38
inc eases in o al c opland han in p oduc ion due o p ojec ed yield g ow h (see deli e able
D6.3).
In a BAU u u e, BRA is expec ed o ha e a ela i ely high and inc easing sha e o land dedica ed
o soy (including bo h single c opping and soy-co n double c opping, wi h only hal alloca ed o
soy o he la e ), and a ela i ely high o es loss a e. Howe e , om 2030 onwa ds, he e is
an inc ease in es o a ion land ollowing he implemen a ion o he Fo es Codes’s Legal Rese e
equi emen s.
In he IAP scena io, pas u e d ops in all egions om bo h 2050 and 2020 BAU le els ollowing
he shi in die s away om animal p oduc s. C opland also declines sligh ly om he 2050 BAU
in all egions (and om 2020 in mos egions), wi h ela i ely la ge decline in land dedica ed o
soy in B azil due o he educed demand o eed c ops and he assumed addi ional yield g ow h.
F eed up ag icul u al land is con e ed o es o a ion land in all egions (as assumed as pa o
he policy package)4, while losses o o es (in BRA, OSA and AME) and o he na u al lands
(mainly in ROW and AME) la gely dec ease compa ed o he baseline (p e en ing he con e sion
o 80Mha o o es and o 56Mha o o he na u al land). A la ge sha e o he es o a ion land a
global scale is p ojec ed o occu in B azil, which is due o a di e en ep esen a ion o pas u e
in B azil as compa ed o o he egions5. The p ojec ed es o a ion in he IAP scena io each
abou 120 million ha, as compa ed o abou 18 million ha es o ed as pa o he Fo es Code in
2030.
In he T ade dis up ion scena io, pas u e (and o es o a lowe ex en ) decline sligh ly in B azil
compa ed o he 2050 BAU scena io while soy a ea inc eases (+4%) o mee he inc ease in soy
impo demand om China.
In he ZNL scena io, pas u e – soy and o he c opland o a lowe ex en - decline in B azil
compa ed o he 2050 BAU scena io, p e en ing he con e sion o 50 Mha o o es and 16Mha
o o he na u al lands. Howe e , he decline in ag icul u al land compa ed o he 2050 BAU is
less p onounced han in he IAP scena io, which also assumes he implemen a ion o o he land
sa ing measu es (i.e., die a y shi , land es o a ion and sus ainable inc ease in c op yields) in
addi ion o conse a ion e o s.
4 In B azil, es o a ion e o s as pa o he IAP a e addi ional o hose as pa o he Fo es Code (wi h la ge inc ease
in es o a ion land compa ed o he baseline).
5 In he s anda d e sion o GLOBIOM, only pas u e land equi ed o eed uminan s (based on assump ions abou
he spa ial dis ibu ion o uminan s, g assland p oduc i i y and g azing in ensi y) is labelled as such, while he
emainde o non- o es and non-c opland ege a ion is classi ied as o he na u al land. This leads o much lowe
amoun s o pas u e land han epo ed by FAOSTAT, and o en assumed in o he land use models (wi h a low
oppo uni y cos , as assumed o ha e e y p oduc i i y). As pas u e is conside ed eligible o es o a ion and no
o he na u al land, GLOBIOM usually p ojec lowe amoun s o es o a ion han o he models. is no conside ed
eligible o es o a ion, he s anda d e sion o GLOBIOM usually p ojec s lowe amoun s o es o a ion han mos
o he models. As o CLEVER, he ep esen a ion o pas u e in made close o ha o o he models (wi h a la ge a ea
iden i ied as pas u e and wi h low oppo uni y cos o es o a ion), amoun s o es o a ion a e la ge o B azil han
o o he egions.
39
Figu e 12. P ojec ed ends in land use (in million hec a es) o i e agg ega ed egions and wo zoomed egions
(B azil BRA and Eu opean Union EUE). The colo s di e en ia e pas u e ( u quoise) om soy c op physical a ea
(pu ple) and o he p ima y c op physical a ea (yellow), as well as o es ( ed), o he land (blue) and es o a ion
land (o ange).
GHG emissions
Figu e 13 shows p ojec ed GHG emissions om li es ock ( h ough en e ic e men a ion and
manu e managemen ), soy and o he c ops ( h ough c opland soil N2O emissions and CH4
emissions om ice cul i a ion), and LUC ( h ough changes in abo e g ound ca bon s ocks) in
he di e en scena ios. O e all, li es ock accoun s o he la ges sha e o AFOLU emissions in
mos egions, excep in BRA and AME whe e LUC emissions domina es.
In he SSP2 scena io, LUC emissions a e p ojec ed o emain s able (when low in 2020) o
dec ease (when high in 2020) in all egions be ween 2020 and 2050, excep in AME whe e an
inc ease is p ojec ed due o he con e sion o unmanaged land o ag icul u al land. C op and
li es ock emissions, in u n, a e p ojec ed o inc ease globally. O e all, o al AFOLU GHG
emissions a e p ojec ed o inc ease in all egions excep in B azil, whe e he decline in LUC
emissions ( ollowing he implemen a ion o he Fo es Code) mo e han o se he inc ease in
di ec emissions om ag icul u e.
In he IAP scena io, GHG emissions dec ease in all egions compa ed o he baseline, mainly due
o a la ge d op in LUC and li es ock emissions, and some decline in c op emissions. LUC becomes
a ca bon sink in mos egions. This decline in AFOLU emissions is due o he assumed shi in
die s away om animal p oduc s (which a e mo e emission in ensi e han c op p oduc s), as
well as he land es o a ion e o s, suppo ed by supply-side and demand-side measu es.
40
In he T ade dis up ion scena io, no signi ican changes in GHG emissions a e p ojec ed
compa ed o he BAU scena io.
In ZNL scena io, B azil’s AFOLU emissions dec ease compa ed o he baseline, mainly due o a
la ge d op in LUC emissions and a small decline in li es ock emissions. Howe e , AFOLU
emissions a e highe han in he IAP scena io, which conside s addi ional measu es on he
demand and supply sides leading o lowe GHG emissions om all sou ces.
Figu e 13. P ojec ed ends in annual GHG emissions (in billion ons o CO2 equi alen s pe yea ) om he
ag icul u e and land use (AFOLU) sec o s o agg ega ed wo ld egions and wo zoomed egions (B azil BRA and
Eu opean Union EUE). The colou s di e en ia e emissions om li es ock p oduc ion ( u quoise, LSP), soy c op
p oduc ion (pu ple), o he c op p oduc ion (yellow), and land use change ( ed).
Wa e use
The p ojec ed changes in wa e use o i iga ed c op p oduc ion a e displayed in Figu e 14. Fo
mos egions, i iga ion wa e use is p ojec ed o emain s able o sligh ly inc ease, wi h
p ojec ed inc eases in wa e use e iciency mi iga ing he inc eases in i iga ed c op p oduc ion
(see deli e able D6.3). Wa e use o soy p oduc ion is only signi ican in NAM (as soy in no
i iga ed in BRA) and is p ojec ed o s ay b oadly s able o e ime.
In he IAP scena io, global wa e use is p ojec ed o be b oadly in line wi h he 2050 BAU, wi h
a decline in some egions (SAS, and AME) and an inc ease in o he s. Soy wa e use in NAM is
p ojec ed o inc ease compa ed o he baseline due o an inc ease in soy p oduc ion (see Figu e
8). In he T ade dis up ion scena io, soy wa e use sligh ly declines om 2050 baseline le els in
41
NAM (-3%) due o a d op in soy p oduc ion ollowing a educ ion in Chinese impo demand (see
Figu e 8). The ZNL scena io has no clea impac on wa e use in B azil.
Figu e 14. P ojec ed ends in wa e use o i iga ion (in km3) o i e agg ega ed egions and wo zoomed egions
(B azil BRA and Eu opean Union EUE). The colou s di e en ia e soy c ops (pu ple) om o he c ops (yellow).
Biodi e si y impac s
Biodi e si y impac s on e es ial ecosys ems om local ag icul u e, measu ed in po en ially
disappea ed ac ion o global species pe yea , a e displayed in Figu e 15 (see D6.3 o mo e
de ails on he me hodology). They a e, in absolu e e ms, highe in egions wi h a signi ican
sha e o opical ecosys ems (e.g., SAS, BRA and OSA, and AME). Land occupa ion has he la ges
impac in all egions, ollowed by land ans o ma ion (excep in NAM whe e clima e change
impac s a e highe ). I should be no ed ha his only accoun s o biodi e si y impac s om he
land use and ag icul u al sec o s. While o land occupa ion and land ans o ma ion impac s
his may be close o o al impac s, signi ican ly la ge clima e change impac s on biodi e si y
would be expec ed i also accoun ing o biodi e si y impac s om o he economic sec o s.
In he BAU scena io, land ans o ma ion impac s a e p ojec ed o decline be ween 2020 and
2050, especially in egions whe e hey a e cu en ly la ge (e.g. BRA and OSA, AME, SAS). Land
occupa ion and clima e impac s, in u n, a e p ojec ed o emain b oadly s able o inc ease o e
ime, excep o clima e impac s in B azil due he implemen a ion o he Fo es Code. I should
be no ed ha , by assump ion, es o a ion does no di ec ly a ec biodi e si y, as only
biodi e si y-de imen al impac s a e accoun ed o .
48
Al hough he T ade dis up ion scena io leads o a sligh edis ibu ion o wa e use om Pampa
o Ma a A lan ica and Ce ado, he p ojec ed le el o wa e use in bo h he T ade dis up ion and
ZNL scena ios do no di e signi ican ly om he BAU scena io.
Figu e 20. P ojec ed ends in Wa e consump ion (wa e o i iga ion, in km3) o he six B azilian biomes. The
colo s di e en ia e he biomes, namely Amazonia (blue), Caa inga (yellow), Ce ado ( ed), Ma aA lan ica
( u quoise), Pampa (g een) and Pan anal (pink).
Biodi e si y impac s
Figu e 21 shows he impac s on e es ial ecosys ems om a ming ac i i ies in he six B azilian
biomes, based on wo di e en LCA me hodologies. In he igu es in he op panel, biodi e si y
impac s a e calcula ed based on he LC-IMPACT a e age cha ac e isa ion ac o s (CFs) (Ve ones
e al., 2020) while hose in he bo om panel a e calcula ed based on he new CFs de eloped o
Sou h Ame ica in he con ex o D2.4 (he ea e e e ed o as ‘CLEVER CFs’) (Oli ei a e al., 2019;
Oli ei a & Pacheco, 2024) I should be no ed ha esul s o his igu e omi clima e change
impac s, which a e a ailable o he LC-IMPACT CFs bu no o he CLEVER CFs.
Bo h me hods p edic simila o e all ends ac oss scena ios. In he BAU SSP2 scena io, wi h
bo h app oaches, he o al land impac s on e es ial ecosys ems a e p ojec ed o decline o e
ime, due o a d op in land ans o ma ion (i.e., LUC) impac s and despi e an inc ease in land
occupa ion impac s. In he IAP scena io, o al impac on e es ial ecosys ems dec eases
compa ed o he baseline, due o a decline in bo h land occupa ion and land ans o ma ion
impac s. In he ZNL scena io, he o al impac on e es ial ecosys ems also declines compa ed
o he BAU mainly due o a d op in land ans o ma ion and some decline in land occupa ion
49
impac s. Howe e , which o he IAP o ZNL scena ios lead o he bes ou come o e es ial
biodi e si y depends on he me hod, wi h he IAP scena io pe o ming be e when using LC-
IMPACT CFs and he ZNL scena io when using he CLEVER CFs. The T ade dis up ion scena io, in
u n, has no signi ican impac on e es ial biodi e si y o bo h me ics.
The main di e ences be ween he wo me hods lie in he scale o he impac s (in pa icula , o
ans o ma ion impac s) and he ela i e con ibu ion o di e en biomes o he o e all impac s.
Fi s , he B azil-le el ans o ma ion impac s o a ming ac i i ies on e es ial ecosys ems a e
mo e han wo imes highe when using he CLEVER CFs han when using he LC-IMPACT CFs (in
line wi h wha was discussed in D6.3). Then, when using he CLEVER CFs, Amazonia is he biome
wi h he la ges o e all impac on e es ial ecosys ems (42% o he o al in 2020), ollowed by
Ce ado (26%) and Ma a A lan ica (17%) while when using he LC-IMPACT CFs, Ma a A lan ica
has he la ges impac (42%), ollowed by Amazonia (31%) and Ce ado (21%).
Finally, he a oided loss o o es and o he na u al land in he ZNL scena io (as compa ed o he
BAU scena io) seem o be equally bene icial o biodi e si y wi h bo h me hods (close o ull
elimina ion o ans o ma ion impac s), while he land use change pa e ns o he IAP scena io
as compa ed o o he scena io seems o esul in much highe bene i s o biodi e si y when
using he LC-IMPACT CFs (close o ull elimina ion o ans o ma ion impac s) han when using
he CLEVER CFs ( e y limi ed dec ease in ans o ma ion impac s). This p o ides impo an
nuances ega ding he bes s a egies o biodi e si y p o ec ion in B azil.
Figu e 21. P ojec ed ends in biodi e si y impac o e es ial ecosys ems (Po en ially disappea ed ac ion,
PDF·y) o he six biomes and spli be ween he di e en p essu es co e ed in GLOBIOM. The colo s di e en ia e
namely Amazonia (blue), Caa inga (yellow), Ce ado ( ed), Ma aA lan ica ( u quoise), Pampa (g een) and Pan anal
(pink). The igu es in he op panel a e based on LC-IMPACT cha ac e isa ion ac o s (CFS) while he igu es in he
bo om panel a e based on spa ially-explici CFs de eloped o Sou h Ame ica in CELEVER deli e able D2.4.
50
As shown in Figu e 22, eshwa e ecosys ems a e mos a ec ed by a ming ac i i ies in he
Ma a A lan ica, Ce ado, and Amazonia biomes.
In he BAU scena io, he o al impac on eshwa e ecosys ems in B azil is p ojec ed o inc ease
be ween 2020 and 2050, due o a ise in eshwa e eu ophica ion in all biomes, and some
inc ease in wa e s ess in some biomes (mainly Ma a A lan ica). Clima e change impac s,
howe e , a e p ojec ed o decline, due o declining LUC emissions.
In he IAP scena io, he impac s on eshwa e ecosys ems a e p ojec ed o be lowe han in he
BAU and become nega i e, mainly due o s ong decline in clima e change impac s (wi h LUC
u ning in o a ca bon sink) and some dec ease in eshwa e eu ophica ion in all biomes.
Howe e , as discussed abo e, wa e s ess sligh ly inc eases om he BAU due o a small
inc ease in wa e use o i iga ed c ops (see Figu e 20).
In he T ade dis up ion scena io, biodi e si y impac s on eshwa e ecosys ems sligh ly inc ease
om he 2050 BAU, mainly due o an inc ease in eshwa e eu ophica ion (and wa e s ess o
a lowe ex en ).
In he ZNL scena io, he o al eshwa e ecosys ems’ impac s a e p ojec ed o be lowe han in
he BAU scena io, mainly due o lowe clima e change and eu ophica ion impac s. Howe e ,
he o al impac on eshwa e ecosys ems emains highe han in he IAP scena io.
Figu e 22. P ojec ed ends in biodi e si y impac o eshwa e ecosys ems (Po en ially disappea ed ac ion,
PDF·y) o he six biomes and spli be ween he di e en p essu es co e ed in GLOBIOM. The colo s di e en ia e
he biomes, namely Amazonia (blue), Caa inga (yellow), Ce ado ( ed), Ma a A lan ica ( u quoise), Pampa (g een)
and Pan anal (pink).
Addi ional impac s om ups eam and downs eam soy supply
chain in B azil
In he p e ious sec ions, he p ojec ed en i onmen al impac s o B azilian ag icul u al ac i i ies
ha e been illus a ed, based on he ou come o GLOBIOM o a BAU scena io (SSP2) and h ee
s ylised policy scena ios. Howe e , pa s o he supply chain o ag icul u al p oduc s a e missing
in GLOBIOM, he model p ima ily co e ing he impac o di ec a ming ac i i ies and ela ed
51
changes in esou ce use. Based on he esul s o he LCA de eloped in D6.4, he model was
expanded o include ups eam and downs eam supply chain impac s o soy p oduc ion in B azil.
These a e es ima ed by combining LCA-based es ima es o impac s pe on o soy p oduced wi h
GLOBIOM p ojec ions o he amoun o soy p oduced. In his sec ion, he addi ional impac s
om hese supply chain s eps a e p esen ed o soy and compa ed o he oo p in om a ming
p esen ed abo e.
We analyse he di e ences in impac in e ms o aqua ic and e es ial ex inc ion isk, bu also
in e ms o GHG emissions and eshwa e consump ion.
Compa ison o a ming impac s o all c op and li es ock p oduc s o soy
impac om a ming and o he supply chain s eps
The in eg a ion o all supply chains impac s in o GLOBIOM o soy allows us o pu hese impac s
in o pe spec i e by compa ing hem o he p e iously calcula ed B azilian impac s om a ming
o all ag icul u al commodi ies (i.e., soy and o he c ops and li es ock p oduc s) in 2020 (Figu e
23). When looking only a he impac s o a ming co e ed in GLOBIOM (in yellow in Figu e 23),
soy p oduc ion alone accoun s o 3% o GHG emissions and aqua ic biodi e si y impac s and
13% o e es ial biodi e si y impac s (soy has no impac o eshwa e consump ion i is no
i iga ed in B azil). This highligh s he ela i e impo ance o soy wi hin all ag icul u al
commodi ies o he a ming impac s co e ed in GLOBIOM.
Howe e , hey a e di e en be ween biomes ega ding he ela i e con ibu ion o soy o he
o al a ming impac s co e ed in GLOBIOM. Fo example, in he Ce ado - whe e 60% o B azil’s
soy p oduc ion akes place - soy p oduc ion accoun s o a la ge sha e o he o al a ming
impac s han a na ional scale o han in any o he biomes. Con e sely, in Amazonia, whe e GHG
emissions and e es ial ecosys ems impac s om a ming a e he la ges , soy p oduc ion
accoun s o a smalle po ion o hese impac s han a he na ional scale.
Figu e 23 also illus a es he con ibu ion o di e en supply chain s eps o soy o e all impac s.
Fo e es ial ecosys em impac s, a ming GLOBIOM accoun s o 90% o he o al impac s,
while o he supply chains s eps a e esponsible o he emaining 10%. Fo aqua ic ecosys ems,
a ming GLOBIOM accoun s o abou hal o he o al impac s, while o he supply chains s eps
(mainly c ushing and inpu p oduc ion) accoun s o he o he hal . Fa ming GLOBIOM has no
impac on eshwa e consump ion as soy is no i iga ed in B azil; mos o he impac s coming
om inpu s p oduc ion (~60%) and c ushing (~30%). Finally, o GHG emissions, a ming impac s
co e ed in GLOBIOM accoun s o abou 25% o he o al impac while o he supply chain s eps
a e esponsible o he emaining 75%. This highligh s he impo ance o conside ing addi ional
supply chain impac s in he s anda d e sion o GLOBIOM, as including hose o o he c ops
would u he inc ease he o e all en i onmen al impac s.
52
Figu e 23. Soy supply chain oo p in compa ed o o al a ming oo p in o B azil and ou soy-p oducing biomes
in 2020 (SSP2). In Yellow, desc ibed as “Fa ming - GLOBIOM” includes all al eady calcula ed impac s om
GLOBIOM, which a e ela ed o di ec ag icul u al and o es y ac i i ies. “Fa ming – o he impac s” mainly includes
impac associa ed wi h inpu p oduc ion such as e ilize .
E olu ion o soy supply chain biodi e si y impac s
When looking a he e olu ion o he biodi e si y impac o soy p oduc ion in B azil (Figu e 24),
we obse e a decline in impac on e es ial ecosys ems o e ime, due o a dec ease in a ming
53
impac s, mo e speci ically in land ans o ma ion, while ups eam and downs eam supply chain
impac s inc ease. The impac on aqua ic ecosys ems, in u n, inc eases o e ime, due o an
inc ease in mos supply chain impac s (excl. a ming), mo e speci ically GHG emissions.
In he BAU scena io, he e is a no iceable inc ease in he biodi e si y impac s on aqua ic
ecosys ems, and GHG emissions a ibu ed o non- a ming supply chain s ages while a ming-
ela ed impac s decline o e ime. I should be no ed ha o o he impac s han he a ming
impac s co e ed in GLOBIOM, he LCA coe icien s used a e es ima ed di ec ly pe on o p oduc
and do no e ol e dynamically wi h he scena ios in GLOBIOM o include po en ial educ ion in
impac s due o echnological p og ess. By con as , a ming impac s co e ed in GLOBIOM used
coe icien pe hec a e o pe on o inpu , and hese impac s a e a ec ed by long- e m inc eases
in yield and inpu use e iciency. Fo o he impac s han he a ming GLOBIOM, and including
aqua ic biodi e si y impac s, his could po en ially lead o an o e es ima ion o he impac s in
he p ojec ions. In con as , he impac s on e es ial ecosys ems, which a e p ima ily d i en by
a ming impac s co e ed in GLOBIOM, closely ollow he ends obse ed o B azil as a whole.
In he IAP scena io, whe e B azil’s soy p oduc ion d ops compa ed o he BAU (see Figu e 17),
he impac s o he soy supply chain on bo h e es ial and aqua ic ecosys ems decline om he
BAU le els (wi h he impac on e es ial ecosys ems becoming nega i e), mainly due o a
dec ease in a ming impac s. Mid-poin s impac s (GHG emissions and eshwa e consump ion)
also decline om he baseline, mainly due lowe impac s om ups eam supply chain ac i i ies
(i.e., inpu p oduc ion) and om c ushing.
In he T ade dis up ion scena io, whe e B azil’s soy p oduc ion sligh ly inc eases om he 2050
BAU, he impac on e es ial ecosys ems is compa able o 2050 baseline le el while he impac
on aqua ic ecosys ems inc eases mainly due o a ise in a ming and ups eam supply chain
impac s. Mid-poin s impac s also inc ease, mainly due o highe impac s om ups eam supply
chain ac i i ies.
In he ZNL scena io, he end-poin impac s (aqua ic and e es ial biodi e si y) o he soy supply
chain a e below 2050 BAU le els mainly due o lowe a ming impac s ollowing a small decline
in B azil soy p oduc ion. Mid-poin s impac s a e also sligh ly below baseline le els, mainly due
o lowe impac s om c ushing o eshwa e consump ion, and om a ming and c ushing in
he case o GHG emissions.
54
Figu e 24. E olu ion o he mid-poin impac s (GHG emissions and eshwa e consump ion) and end-poin impac s
( eshwa e and e es ial biodi e si y) o he soy supply chain o e ime, di e en ia ed pe supply chain s ep.
Discussion
Soy is one o he mos in e na ionally aded ag icul u al commodi ies. O e he pas decades,
i s p oduc ion and ade expanded apidly, mainly d i en by s ong impo demand o p o ein
meals in China and he Eu opean Union. These in e na ional demands ha e been associa ed
wi h land use change, de o es a ion, and subsequen biodi e si y loss in B azil, he wo ld’s
la ges p oduce and expo e o soybeans. In his deli e able, we explo ed al e na i e u u e
de elopmen s in he soy ma ke s, wi h a ocus on B azil and he EU.
T ends in a business as usual u u e
When conside ing he con inua ion o his o ical ends in popula ion, die s, ade and
p oduc i i y, and no change in cu en policies, he global p oduc ion, consump ion and ade
o li es ock and c op p oduc s a e p ojec ed o con inue inc easing by 2050. This includes
con inuing g ow h in soy p oduc ion and ade in soy-based p oduc s, wi h B azil p ojec ed o
accoun o mo e han hal o he global g ow h in ou pu and ne expo s. Mos o he inc ease
in expo s a e des ina ed o Asia, while EU also inc eases i s le el o impo s. These b oad
p oduc ion and consump ion pa e ns a e associa ed wi h g owing en i onmen al p essu es,
leading o u he clima e change impac s and inc easing ex inc ion isk o aqua ic and
e es ial ecosys ems, in pa icula in La in Ame ica, Asia and A ica.
In B azil, soy p oduc ion inc ease is p ojec ed o occu p ima ily in he Ce ado biome, and o
some ex en , in he Ma a A lan ica and Pampa biomes. This g ow h is achie ed in la ge pa s by
a con e sion o pas u es and yield inc eases. De o es a ion is p ojec ed o con inue un il 2050
albei a a slowe pace han in p e ious decades. La ge o es con e sions o pas u e a e
p ojec ed, in pa icula in he Amazon biome, while mode a e o es es o a ion e o s ake
place, in pa icula in he Ma a A lan ica biome. I should be no ed ha hese ends conside
he e ec s om key domes ic in e en ions in B azil, such as he Amazon Soy Mo a o ium
(limi ing soy expansion in he Amazon Biome) and he Fo es Code (limi ing bu no ully
elimina ing illegal de o es a ion, and igge ing mode a e es o a ion e o s). This leads o a
55
small educ ion in GHG emissions om he AFOLU sec o , p ima ily d i en by a dec ease in land
use change emissions pa ially compensa ed by an inc ease in GHG emissions om li es ock
p oduc ion. Concomi an ly, inc eases he alue o p oduc ion a e p ojec ed, in pa icula o soy,
bu hese emain o a much lowe ampli ude han p ojec ed in inc eases in p oduc ion olume.
Ex inc ion isks om ag icul u al ac i i ies inc ease o aqua ic ecosys ems o B azil, in pa icula
due o eu ophica ion in he Ce ado biome. Fo e es ial ecosys ems, impac s om ag icul u e
sligh ly decline o hose associa ed wi h GHG emissions and land use change ( e lec ing he
slowing a es o de o es a ion and es o a ion e o s), bu inc ease o impac s associa ed wi h
land occupa ion (indica ing ha he land sec o is s ill leading o ne inc eases in land-use change
media ed e es ial biodi e si y loss).
Al hough a de ailed compa ison o he li e a u e would be ele an o aspec s such as China’s
and EU’s u u e demand o soy-based p oduc s, hese ends a e b oadly consis en a global
scale wi h he li e a u e o u u e p ojec ions o business as usual scena ios (e.g., o SSP2
scena io (Popp e al., 2017)) and medium- e m ag icul u al ou looks (OECD/FAO, 2024), and
simila s udies a he scale o B azil (e.g., (So e oni e al., 2018)). I migh be ele an o conside
addi ional global d i e s, such as impac s om clima e change, bu hese a e no expec ed o
lead o la ge changes in B azil’s soy expo po en ial (Zilli e al., 2020).
Impac o al e na i e u u es
We also designed explo a i e scena ios cap u ing key sou ces o unce ain y o soy ma ke s (as
iden i ied in he li e a u e and du ing a s akeholde wo kshop) and analysed hei socio-
economic and en i onmen al impac s a ound he BAU, wi h a ocus on hei implica ions o
BRA-EU soy supply chains. This includes scena ios ep esen ing a) a global ood sys em
ans o ma ion (i.e., he IAP scena io), b) he po en ial long- e m implica ions o he ade
dispu e be ween US and China (i.e., he T ade dis up ion scena io), and c) an idealised ambi ious
land conse a ion policy in B azil (i.e., he ZNL scena io).
Global ood sys em sus ainabili y ansi ion
The IAP scena io assumes he global implemen a ion o a mix o demand-side (incl. die a y shi ,
was e educ ion), supply-side (e.g., sus ainable yield inc eases) and conse a ion (inc eased
p o ec ed a ea ex en and e ec i eness, land use planning) and es o a ion measu es. In his
scena io, he p ojec ed global p oduc ion and ade o li es ock p oduc s and eed c ops (incl.
soy) by 2050 d ops compa ed o he BAU. This leads o a much mo e limi ed inc ease in ne
expo s o soy- ela ed p oduc s om B azil (+20% ins ead o +76%). Ne expo s o soy-based
p oduc s o he EU dec ease compa ed o 2020 le els, while mos o he impo ing egions
main ain o al impo s le els sligh ly abo e 2020 le els. This scena io is p ojec ed o lead o
decline in all en i onmen al impac s bu also en ail impo an socio-economic ade-o s, wi h
la ge dec eases in he alue o li es ock p oduc ion, p ima ily as a esul o changes in consume
p e e ences.
In B azil, a signi ican amoun o pas u e is es o ed, while o es and o he na u al land losses
a e pa ially mi iga ed, as compa ed o he BAU scena io. Land-use change becomes a la ge
56
ca bon sink, and he biodi e si y impac s o local ag icul u e on bo h aqua ic and e es ial
ecosys ems a e la gely educed. Howe e , i leads o signi ican o gone economic oppo uni ies
o a me s, wi h he alue o ag icul u al p oduc ion declining om 2050 bu also om 2020
BAU le els. Dec eases a e pa icula ly la ge (mo e han -50% as compa ed o 2020 le els) o
li es ock p oduc ion. Dec ease a e e y mode a e o o he c ops excep o OSA, while o soy,
and in pa icula soy in B azil, he alue o p oduc ion dec ease compa ed o 2020 le els.
Ambi ious conse a ion in B azil
En o cing ze o con e sion o o es and o he na u al lands o ag icul u e in B azil om 2020 (as
assumed in he ZNL scena io) only leads o a sligh ly lowe pos -2020 g ow h in he p oduc ion
and ade o li es ock and soy p oduc s compa ed o he BAU, and he e o e on ag icul u al and
land use ends in he es o he wo ld. The hal ing o land con e sion leads o a d op in LUC
emissions, and educed biodi e si y impac s on bo h aqua ic and e es ial ecosys ems
compa ed o he BAU. Howe e , biodi e si y impac s emain highe han in he IAP scena io,
which also conside s land es o a ion and supply and demand-side measu es. On he o he
hand, he alue o p oduc ion is p ojec ed o no be signi ican ly a ec ed in he ZNL scena io, in
con as o he s a k declines p ojec ed in he IAP scena io, p ima ily in he li es ock sec o , bu
also in he soy sec o .
This sugges s ha B azil could keep supplying domes ic and wo ld ma ke s wi hou clea ing
o es and o he na u al lands, he eby pu suing bo h en i onmen al and economic goals, in
pa icula h ough mobilizing pas u e o soy p oduc ion. Howe e , he same pas u e land is also
p ojec ed o be es o ed in he scena io pu suing ambi ious biodi e si y goals, wi h clea
bene i s in e ms o clima e mi iga ion. In such a scena io, he a e o B azilian pas u es he e o e
appea s a he cen e o po en ial con lic s be ween economic and en i onmen al goals, while
economic oppo uni ies loss would eme ge om a shi in global consump ion pa e ns, wi h
po en ially la ge implica ions o li es ock p oduce s. The e migh be a oom o scena ios wi h
mo e mode a e changes in consump ion pa e ns in B azil, and a mix o es o a ion and
sus ainable in ensi ica ion o he li es ock sec o (Cohn e al., 2014; De Oli ei a Sil a e al.,
2018).
Long- e m impac s o he US-China ade dispu e
As compa ed o o he al e na i e scena ios, he long- e m impac s he US-China ade dispu e
migh be limi ed. In his scena io we assume he ade dispu e o esul in a long- e m dec ease
in US soy expo s o China, capped a maximum 75% o hei 2020 le els. This scena io leads o
some inc ease in soy p oduc ion and ade in B azil compa ed o he 2050 BAU, a pa e n e y
simila o he sho - e m impac s obse ed in esponse o he ade shocks in he ecen yea s.
Howe e , we p ojec ed his scena io o ha e no impac on de o es a ion, wi h an only sligh ly
highe inc ease soy a ea (mos ly a he expense o pas u e in he Ce ado biome) o e 2020-
2050 as compa ed o he BAU scena io. We ound no signi ican impac o his scena io on B azil-
le el p ojec ed biodi e si y impac s on aqua ic and e es ial ecosys ems as compa ed o he
BAU scena io.
57
In e es ingly, we ound he sho all in USA soya expo s o China o be edis ibu ed o o he
expo des ina ion only o a e y limi ed ex en , and he B azil expo s o o he des ina ions han
China o be e y li le a ec ed. Al hough la ge e o s om he USA o de elop al e na i e soya
expo ma ke s could be assumed, his e lec s ha o e all he in e na ional ma ke o soy-
based p oduc s o become e y compe i i e and al hough sligh ly inc easing i s su plus, he USA
is p ojec ed o become a less impo an playe . In such a con ex , i will be di icul o soy
p oduce s in he USA o eco e om sho - e m d ops in expo oppo uni ies, which migh
en ail long- e m isks.
I should be acknowledged ha he e is conside able unce ain y in he long- e m impac s om
he US-China ade dispu e, and a b oade ange o assump ions migh be wo h es ing. I
should also be no ed ha ou esul s e lec wo di e en aspec s o he baseline scena io: i s ,
soy expo s om B azil, in pa icula o China, a e p ojec ed o al eady massi ely inc ease in he
baseline scena io, while expo s om he USA o China a e p ojec ed o unde go limi ed
inc eases. This makes bo h he assumed China soya impo sho all, and he po en ial addi ional
expo oppo uni ies o B azil soya bean o China, limi ed by 2050 in ela i e e ms. Second,
his assumes ha ha al hough only pa ially limi ing de o es a ion, domes ic in e en ions in
B azil such as he Amazon Soy Mo a o ium and he Fo es Code a e in place. Should his no be
he case, he soy p oduc ion and expo de elopmen in B azil lead o highe a es o
de o es a ion in scena ios wi h high demand o soya expo s.
A ich pic u e on unce ain u u e biodi e si y impac s
In his deli e able, we ely on mul iple indica o s o quan i y biodi e si y impac s, elying on
connec ing a ious LCIA-based cha ac e iza ion ac o s o a ious ou pu s he GLOBIOM model
(see deli e ables D6.3 and D6.4). This included p e-exis ing cha ac e iza ion ac o s om he LC-
IMPACT p ojec , as well as new cha ac e iza ion ac o s gene a ed in he CLEVER p ojec and
ailo ed o a mo e accu a e ep esen a ion o he B azilian con ex . These indica o s enable us
o ansla e esou ce use (land, wa e ), inpu use and GHG emissions (as quan i ied in GLOBIOM)
in o biodi e si y impac s on bo h aqua ic and e es ial ecosys ems, h ough mul iple impac
pa hways (e.g., land use, wa e use, clima e change, nu ien pollu ion).
This p o ides a signi ican ad ancemen o he li e a u e, wi h he ew land use models
connec ed o biodi e si y models o en ocusing on e es ial ecosys ems and land use change
only (e.g., (Leclè e e al., 2020)), and in a e cases including a b oade se o impac s (e.g.,
(Schippe e al., 2020)). We o example explo ed he impac o u u e soy supply chains
de elopmen s on eshwa e ecosys ems, wi h an impo an ole o wa e use ( o soy p oduced
in he USA) and nu ien losses and GHG emissions ( o soy p oduced in B azil). This allowed us
enable us o cap u e, o B azil soy supply chains, he impac s o supply chain s eps no co e ed
by he GLOBIOM model, bo h ups eam (e.g., GHG emissions om e ilize p oduc ion) and
downs eam (e.g., domes ic and in e na ional anspo o soy commodi ies) o he a ming
s age. Fo clima e change, o ins ance, a ming accoun s o abou 25% o he o al impac while
o he supply chain s eps (mainly inpu p oduc ion and c ushing) a e esponsible o he
emaining 75%. Mo eo e , o aqua ic ecosys ems and clima e change impac s, he sha e
64
land (AbdLnd) a he han biodi e si y ich na u al land (Na Lnd) (Figu e 30). This implies ha
he di ec land-use change impac o plan a ions expansion on biodi e si y emains small and
he indi ec land-use change impac is posi i e (i.e., spa ed na u al o es s a ea).
Figu e 29. O e all land-use change (Mha)
Figu e 30. Plan a ions (ene gy c ops + o es s) expansion and dec ease in abandoned s. na u al land a ea (Mha)
0
500
1000
1500
2000
2500
3000
Mha
BASE
0
500
1000
1500
2000
2500
3000
Mha
PLA_BIO&CON 2000
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
100
150
200
250
300
350
400
2020 2030 2040 2050 2060 2070 2080 2090 2100
Plan a ions
0
50
100
150
200
250
300
2020 2030 2040 2050 2060 2070 2080 2090 2100
Abandonend land
2800
2810
2820
2830
2840
2850
2860
2870
2880
2890
2900
2000 2020 2040 2060 2080 2100
Na u al land BASE
BIO
CON
BIO&CON
PLA
PLA_BIO
PLA_CON
PLA_BIO&CON
65
Fo es ca bon balance
In he BASE scena io, o es ca bon s o age inc eases o e ime (Figu e 31). The main eason is
a o es a ion up ake, which compensa es o ca bon s o age dec ease in he exis ing o es a ea.
The exis ing o es a ea is a sink un il 2050, bu a e 2050 i u ns o a sou ce due o inc eased
ha es olumes, na u al mo ali y and he sa u a ion o biomass g ow h. In addi ion o
a o es a ion, he ca bon changes in he exis ing o es a ea a e compensa ed by HWP and
BECCS. Howe e , he impac o HWP and BECCS on o e all ca bon balance is ela i ely small in
he BASE scena io due o “business as usual” bioeconomy de elopmen .
Figu e 31. Global o es ca bon s ock and ca bon sink/sou ce in he BASE scena io
In he CON scena io, o es ca bon s o age emains lowe han in he BASE scena io (Figu e 32).
The eason is highe ha es olumes, which dec eases ca bon in he exis ing o es s mo e han
is compensa ed by he inc ease in HWP ca bon s o age.
0
50
100
150
200
250
300
350
400
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon s o age (PgC)
BECCS
HWP
A o es a ion
Exis ing o es s
-6000
-4000
-2000
0
2000
4000
6000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon sink/sou ce (M CO2/y ) De o es a ion
Exis ing o es
managemen
A o es a ion
HWP
BECCS
o al
66
Figu e 32. Global o es ca bon s ock and sink/sou ce in he CON scena io
In he BIO scena io, o es ca bon s o age is highe han in he BASE and CON scena ios (Figu e
33). The e a e wo easons o his. Fi s , he highe bioene gy demand does no inc ease
ha es s and ca bon emissions as much as he highe wood-based cons uc ion ma e ials
demand, because he majo i y o addi ional bioene gy demand is sa is ied by ene gy c ops.
Second, he highe bioene gy demand inc eases ca bon s o age o BECSS conside ably
compa ed o “business-as-usual” bioene gy demand, especially a e 2050 when la ge scale
implemen a ions o BECCS becomes easible due o high ca bon p ices.
0
50
100
150
200
250
300
350
400
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon s o age (PgC)
BECCS
HWP
A o es a ion
Exis ing
o es s
-6000
-4000
-2000
0
2000
4000
6000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon sink/sou ce (M CO2/y ) De o es a ion
Exis ing o es
managemen
A o es a ion
HWP
BECCS
o al
67
Figu e 33. Global o es ca bon s ock and sink/sou ce in he BIO scena io
In he PLA CON scena io, o es ca bon s o age is highe han in he BASE and CON scena ios bu
no as high as in he BIO scena io (Figu e 34). The e a e wo easons o his. Fi s , addi ional
ha es s come om o es s plan a ions ins ead o na u al/semina u al o es s, which keeps he
exis ing o es a ea ca bon emissions a he BASE scena io le el. Second, HWP s o age is highe
han in he BASE scena io.
0
50
100
150
200
250
300
350
400
450
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon s o age (PgC)
BECCS
HWP
A o es a ion
Exis ing
o es s
-6000
-4000
-2000
0
2000
4000
6000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon sink/sou ce (M CO2/y ) De o es a ion
Exis ing o es
managemen
A o es a ion
HWP
BECCS
o al
68
Figu e 34. Global o es ca bon s ock and sink/sou ce in PLA CON scena io
To unde s and he ole o HWP and BECCS in o es ca bon balance, we compa e o es ca bon
s ock and sink wi h and wi hou HWP and BECCS. In gene al, he o es ca bon s o age is
inc easing in all scena ios. Wi hou HWP and BECCS, howe e , he inc ease s a s o sa u a e
a e 2050 (Figu e 35). Adding HWP and BECCS in he o es ca bon s o age imp o es o es
ca bon balance in all scena ios and ends o elimina e he sa u a ion (Figu e 36). The eason is
ha HWP and BECCS emo e old ees and make space o new ees and/o emaining ees
eg ow h. This has basically a simila e ec han na u al mo ali y, bu he di e ence is ha HWP
and BECCS s o e he ca bon o he emo ed biomass ins ead o le ing i o decompose and emi
in he a mosphe e.
0
100
200
300
400
500
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon s o age (PgC)
BECCS
HWP
A o es a ion
Exis ing
o es s
-8000
-6000
-4000
-2000
0
2000
4000
6000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Ca bon sink/sou ce (M CO2/y ) De o es a ion
Exis ing o es
managemen
A o es a ion
HWP
BECCS
o al
69
Figu e 35. Fo es ca bon s o age (PgC) and sink (M CO2/y ) including only he biogenic ca bon ha s ays in
o es s (F)
Figu e 36. Fo es ca bon s o age (PgC) and sink (M CO2/y ) including biogenic ca bon ha s ays in o es s (F) and
biogenic ca bon ha is s o ed ou side o es s (HWP and BECCS)
Biodi e si y
Biodi e si y loss is measu ed as in D6.2 by he Po en ial Disappea ed F ac ion o global species
agg ega ed o e all di e en axa (PDF %). Biodi e si y loss is lowe in he PLA scena ios han in
he “business as usual” o es managemen scena ios (). In pa icula , biodi e si y loss is highes
in he CON and BIO&CON scena ios, whe e inc easing demand o wood-base cons uc ion
ma e ials leads o na u al o es s ex ac ion o p oduc ion. In Figu e 38, biodi e si y loss is
di ided in o di e en land-uses. I shows he impac o na u al o es s in he BIO&CON scena io
is clea ly highe han in he PLA BIO&CON scena io. On he o he hand, he plan a ion
290
310
330
350
370
390
410
2000 2020 2040 2060 2080 2100
Ca bon s o age F
-6000
-5000
-4000
-3000
-2000
-1000
0
2020 2040 2060 2080 2100
Ca bon sink F
BASE
BIO
CON
BIO&CON
PLA
PLA_BIO
PLA_CON
PLA_BIO&CON
290
310
330
350
370
390
410
2000 2020 2040 2060 2080 2100
Ca bon s o age
F+HWP+BECCS
-6000
-5000
-4000
-3000
-2000
-1000
0
2020 2040 2060 2080 2100
Ca bon sink
F+HWP+BECCS
BASE
BIO
CON
BIO&CON
PLA
PLA_BIO
PLA_CON
PLA_BIO&CON
70
expansions in he PLA BIO&CON scena io causes less biodi e si y loss han na u al o es s
ha es s in he BIO&CON scena io.
Figu e 37. Global biodi e si y loss
Figu e 38. Global biodi e si y loss di ided in o di e en land-uses
0
2
4
6
8
10
12
14
16
18
20
22
BASE
0
2
4
6
8
10
12
14
16
18
20
22
BIO&CON
0
2
4
6
8
10
12
14
16
18
20
22
PLA BIO&CON
Plan a ions
Na u al o es s
G assland
C opland
14
15
16
17
18
19
20
21
22
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
To al biodi e si y BASE
BIO
CON
BIO&CON
PLA
PLA_BIO
PLA_CON
PLA_BIO&CON
71
Discussion
The p oblem o he wood-based bioeconomy is ha i ends o inc ease ha es s olumes
conside ably ela i e o cu en ha es olumes, which migh ha e nega i e impac s on o es
ca bon balance and biodi e si y. He e we conside he possibili y o a oiding hese e ec s by
shi ing logging om na u al o es s o plan a ions o es s. Cu en ly plan a ion o es s co e
abou 10% o he global p oduc ion o es a ea and accoun o abou 30% o he global
oundwood supply. Acco ding o ou esul , he nega i e e ec s o he bioeconomy on o es
ca bon balance and biodi e si y could be a oided i we would double he plan a ion o es s a ea
and p o ide 60% o he global oundwood supply om hese plan a ions.
The ques ion i we should p oduce woody biomass in na u al/semina u al o plan a ion o es s
is analogues o he land-sha ing s. land-spa ing discussion in conse a ion ecology and
ag icul u al landscape managemen (Salles e al., 2017). Land-sha ing combines p oduc ion and
na u e conse a ion in he same a ea while land-spa ing sepa a es hese objec i es in di e en
a eas, which implies ha ou cu en managemen scena ios can be in e p e as land-sha ing
while ou plan a ion o es s scena ios as land-spa ing. Acco ding o p e ious s udies ocusing
on o es landscape managemen , land-sha ing ends o wo k be e in egions wi h a long
his o y o o es managemen , low biodi e si y and slow biomass g ow h while land-spa ing
wo ks be e in e ol ing o es indus y egions wi h high biodi e si y and as biomass g ow h
(Be s e al., 2021).
Ou s akeholde s emphasized he ole o expanding mul i-pu pose o es managemen s, which
would en ail mo e na u al/semina u al o es land managed o maximizing mul iple ecosys em
se ices a he han wood p oduc ion, in he di ec ion o a land-sha ing app oach. Acco ding o
ou scena io esul s, land-sha ing implies ha mo e na u al/semina u al o es s a ea is needed
o p oduc ion use, which ends o inc ease exis ing o es a ea ca bon emissions and
biodi e si y loss. This e ec is signi ican ly la ge in high cons uc ion ma e ials demand scena io
han in he high bioene gy demand scena io, because mos o he bioene gy inc ease is me by
an inc ease in ene gy c ops p oduc ion a he han in oundwood ha es ing. Inc eased ca bon
emissions can be compensa ed by highe HWP and BECCS ca bon s o ages. The BECCS ca bon
s o age is ypically highe han he HWP ca bon s o age, since HWP p o ides a empo a y ca bon
s o age while BECCS a pe manen ca bon s o age. On he o he hand, BECCS is a new and
ela i ely expensi e echnology, and becomes economically a ailable only a e 2050 when
ca bon p ices a e high enough.
Land spa ing implies ha less na u al/semina u al o es s a ea is needed o p oduc ion use,
which ends o dec ease exis ing o es a ea ca bon emissions and biodi e si y loss. The spa ed
na u al o es a ea could be used o ca bon seques a ion and na u e conse a ion, as also
demanded by s akeholde s. On he o he hand, unde he land spa ing scena io, mo e land is
needed o plan a ions, which could po en ially dec ease ca bon seques a ion and biodi e si y,
as plan a ions ca bon s o age and biodi e si y a e low. This p oblem is a oided by loca ing new
plan a ions mos ly in abandoned land a he han biodi e si y and ca bon ich na u al o es s o
na u al land.
72
Aquacul u e & aqua eed supply chains
Resul s
Demand o inal blue ood p oduc s
P ojec ions o he inal demand o blue ood p oduc s a e p esen ed in Figu e 39, wi h di e en
ime ho izons o he BAU scena io (2010, 2020, 2030, 2040 and 2050), and 2050 alues o o he
scena ios.
In he BAU scena io, he global demand o ish inal p oduc s emains ela i ely s able o sligh ly
inc eases (i.e., salmonoids SALM, less han +15%) be ween 2020 and 2050 o mos p oduc s,
wi h a ew excep ions. On he one hand, some p oduc s a e p ojec ed o unde go a mo e
signi ican inc ease in global demand, o example eshwa e (FRSH, +51%) and sh imps &
p awns (SHRI, +37%), e en hough hei demand inc ease a lowe speed han in ecen decades
(e.g., +51% o e 2020-2050 s +41% o e 2010-2020). This is compa ible wi h p ojec ions om
o he s s udies (e.g., Naylo e al 2021), wi h he le el o wild ca ch expec ed o s abilize.
Coun ies om Eas e n Asia (EAS, including China) a e expec ed o emain he la ges consume s
o eshwa e and c us acean p oduc s and unde go mode a e demand g ow h, while coun ies
om Sou h (SAS, e.g., India), Sou h-Eas Asia (SEA) o Sub-Saha an A ica seeing a la ge ela i e
inc ease by 2050. Fo some wild ca ch-based ish p oduc s (e.g., pelagic ishes PELG, deme sal
ishes DMRS, ma ine ishes MARN, unas TUNA), he globally s able demand hides mo e
con as ed changes a he egional le el, wi h demand o en inc easing in Sub-Saha an A ica
(SSA) and dec easing in Eas e n Asia (EAS). On he o he hand, demand o ish meal and ish oil
a e expec ed o dec ease subs an ially o e he 2020-2050 pe iod, by abou 90% and 75%,
espec i ely. This ollows he assumed p olonga ion o ecen ends in ed aquacul u e in e ms
o eed con e sion e iciency gains and subs i u ion o ish meal and ish oil eed by c op-based
eed. Mos o he ends in inal p oduc s a e ela ed o ood use and o some ex en o he uses,
excep o ish meal and ish oil, ha a e used o eed.
In o he scena ios, he ends ollow he BAU scena io, wi h a ew excep ions. Fi s , in he BFS20
coun e ac ual scena io, demand emains by design cons an a 2020 le els a e 2020. Second,
in he UNFEDAC20 sensi i i y scena io, he capaci y o un ed aquacul u e sys ems is assumed o
emain cons an a 2020 le els a e 2020, leading o lowe demand by 2050 o some p oduc s
( eshwa e ishes FRSH, c us aceans CRST, sh imps and p awns SHRI). The di e ence o he BAU
demand by 2050 o hese p oduc s e lec s he deg ee o which hey ely on un ed aquacul u e
sys ems: high o CRST (2050 demand is close o 2020 le els and well below BAU le els),
mode a e o FRSH and SHRI (2050 demand is close o BAU le els and well abo e 2020 le els).
Las ly, as opposed o he BAU scena io, he demand o ish meal (FSHM) and ish oil (FSHO)
inc eases o highe le els han 2020 o he AF20 and AFCOMPO20 sensi i i y scena ios. This
esul s om he assumed cons an sha e o ish meal and ish oil in aqua eed equi emen s pe
uni o ou pu ( o 2020 le els, o bo h scena ios), and cons an o al aqua eed equi emen s
pe uni o ou pu ( o 2020 le els, o he AF20 scena io). This highligh s he impo ance o he
expec ed p olonga ion o ecen ends in aquacul u e eeding p ac ices (in e ms o e iciency
and aqua eed composi ion) in shaping u u e demands o ish-based aqua eed, in a con ex o
g owing demand o aquacul u e p oduc s.
73
Figu e 39 - P ojec ions o demand o blue ood inal p oduc s o ood, eed and o he use (i.e., excluding inpu o
educ ion sec o ) by p oduc , agg ega ed o en wo ld egions.
P ima y blue ood p oduc supply and aqua eed use
Global-scale p ojec ions o p ima y ish p oduc s supply by sou ce (ca ch, ed and un ed
aquacul u e) and c op aqua eed equi emen s (by c op) a e p esen ed in Figu e 40. By
assump ion, mos o he supply inc ease is p ojec ed o be sou ced om ed aquacul u e. In he
BAU scena io, he o al supply o p ima y blue ood p oduc s is expec ed o con inue inc easing,
bu a a slowe pace han in ecen pas . I inc eases om 169 million ons (M ) in 2020 o 210
M by 2050 (+24%, oughly equal o he ela i e inc ease om 2010 o 2020). This ansla es
in o a 57% inc ease in ed aquacul u e supply ( om 50 M in 2020 o 78 M in 2050) and a 45%
inc ease in un ed aquacul u e ( om 29 M in 2020 o 41 M in 2050). These ends emain below
( o ed aquacul u e) o nea ing ( o un ed aquacul u e) 2010-2020 a es o inc ease (+87% o
ed aquacul u e, +18% o un ed aquacul u e). These p ojec ions a e quali a i ely compa able
o hose om he OECD-FAO Ou look 2024-2033 (OECD-FAO 2024).
P ojec ions o p ima y ish supply by 2050 o o he scena ios a e simila o ha o he BAU,
excep o wo scena ios. Fo he BFS20 coun e ac ual scena io, supply emains by assump ion
cons an a 2020 le els o all h ee sou ces. Fo he UNFEDAC20 sensi i i y scena io, un ed
aquacul u e supply emains by assump ion cons an a 2020 le els, leading o a sligh ly highe
inc eases in ed aquacul u e (+59% o e 2020-2050, ins ead o +57% in he BAU scena io),
80
p oduc ion in 2020, and his sha e is no expec ed o inc ease by 2050, unless he u u e changes
in eeding p ac ices assumed in he BAU do no ma e ialize. O e he same pe iod, we p ojec
mo e han 300 million hec a es o o es and non- o es na u al land o be con e ed o
ag icul u e, wi h e y li le a ia ion ac oss scena ios, sugges ing ha u u e ends in he blue
ood sec o will ha e limi ed impac on land use change a he global scale, and ha he speci ic
impac s o u u e aqua eed equi emen s maybe challenging o cap u e. We could isola e hese
speci ic impac s by using a coun e ac ual scena io in which he blue ood sec o (demand,
supply, eeding p ac ices) emains cons an a 2020 le el, and in es iga e he ole o a ious
assump ions abou u u e ends in he blue ood sec o by compa ing impac s be ween he
coun e ac ual and no only he baseline scena io, bu also o he sensi i i y scena ios.
We ound he impac s on global e es ial species ex inc ion isks om land occupa ion
a ibu able o u u e c op aqua eed equi emen s o be nega i e bu mode a e, and ou esul s
illus a e a leas ou di e en sou ces o complexi ies when es ima ing hose. Fi s ,
assump ions abou u u e ends in aquacul u e eeding p ac ices such as o e all e iciency and
sha e o c op- s. ish-based aqua eed will modula e he u u e inc ease in o al c op aqua eed
demand, wi h some o hese ac o s (e.g., inc eased o e all e iciency s. inc eased sha e o
c op-based p oduc s in o al aqua eed) expec ed o play in opposi e di ec ion. Second, he
ela ionship be ween na u al land loss and biodi e si y loss is a iable, due o spa ial di e ences
in biodi e si y pa e ns no only ac oss bu wi hin egions. Fo example, he amoun o global
species ex inc ion isks ela i e o na u al land loss was sys ema ically highe o La in Ame ica
and Ca ibbean, as well as Sou h Asia and Sou h-Eas e n Asia, han o o he wo ld egions, and
his link can a y ac oss scena ios. Thi d, ade dependencies migh displace he ela ed inc ease
in c op p oduc ion o biodi e si y ho spo s, o an ampli ude ha may change wi h change wi h
u u e ade pa e ns. Fo example, while less han 10% o he es ima ed u u e inc eases in c op
aqua eed equi emen s is p ojec ed o be loca ed in La in Ame ica and he Ca ibbean, his egion
was ound o hos a leas a qua e o global a ibu able na u al land loss in mos scena ios, in
ela ion o he demand o c ops like soya and co n. Fou h, assump ions abou he composi ion
o c op-based aqua eed equi emen s, whose u u e pa e ns a e gene ally no well cons ained,
can ha e la ge applica ions. Fo example, we ound ha eplacing hal o co n and soya c op
aqua eed demand by c ops like whea (wi h high p o ein con en and g own in empe a e
egions) in China could signi ican ly educe isks o biodi e si y loss in La in Ame ica and he
Ca ibbean.
A las , we also ound he p ojec ed de elopmen s o in c op aqua eed equi emen s o gene a e
addi ional nu ien losses on in ish aquacul u e a ms, in pa icula o eshwa e aquacul u e
in Eas e n Asia. Simila pa e ns migh be expec ed o non- in ish aquacul u e, and losses a e
also expec ed o phospho ous nu ien s, poin ing o bo h inc eased eu ophica ion isks
associa ed wi h aquacul u e de elopmen s, and bene i s o in eg a ed aquacul u e sys ems wi h
nu ien ecycling (e.g., (Xiao e al., 2017)). In addi ion, he inc eased demand o c op aqua eed
is likely o gene a e addi ional nu ien losses wi hin c opland a ms, which migh be possible o
quan i y wi h ou modelling amewo k.
81
PROJECT OUTPUTS ACHIEVED
Lis o Zenodo eposi o ies con aining model p ojec ions:
• Zenodo eco d o soy supply chains: h ps://zenodo.o g/ eco ds/15875871
• Zenodo eco d o o es supply chains: h ps://zenodo.o g/ eco ds/15829433
• Zenodo eco d o aquacul u e & aqua eed supply chains:
h ps://zenodo.o g/ eco ds/15875624
82
REFERENCES
Azue o-Ped aza, C., Lau i, P., Lessa De ci Augus ynczik, A., & Thomas, V. (2024). Managing
Fo es s o Biodi e si y Conse a ion and Clima e Change Mi iga ion.
h ps://pubs.acs.o g/doi/10.1021/acs.es .3c07163
Besna d, S., Koi ala, S., San o o, M., Webe , U., Nelson, J., Gü e , J., He aul , B., Kassi, J.,
N’Guessan, A., Neigh, C., Poul e , B., Zhang, T., & Ca alhais, N. (2021). Mapping global
o es age om o es in en o ies, biomass and clima e da a. Ea h Sys em Science
Da a, 13(10), 4881–4896. h ps://doi.o g/10.5194/essd-13-4881-2021
Be s, M. G., Phalan, B. T., Wol , C., Bake , S. C., Messie , C., Pue mann, K. J., G een, R., Ha is,
S. H., Edwa ds, D. P., Lindenmaye , D. B., & Balm o d, A. (2021). P oducing wood a
leas cos o biodi e si y: In eg a ing T iad and sha ing–spa ing app oaches o in o m
o es landscape managemen . Biological Re iews, 96(4), 1301–1317.
h ps://doi.o g/10.1111/b .12703
Bond, W. (2016). Ancien g asslands a isk.
h ps://www.science.o g/doi/10.1126/science.aad5132
B us, D. J., Henge eld, G. M., Wal oo , D. J. J., Goedha , P. W., Heidema, A. H., Nabuu s, G. J.,
& Gunia, K. (2012). S a is ical mapping o ee species o e Eu ope. Eu opean Jou nal
o Fo es Resea ch, 131(1), 145–157. h ps://doi.o g/10.1007/s10342-011-0513-5
Cabe na d, L., P is e , S., & Hellweg, S. (2024). Biodi e si y impac s o ecen land-use change
d i en by inc eases in ag i- ood impo s. Na u e Sus ainabili y, 7(11), 1512–1524.
h ps://doi.o g/10.1038/s41893-024-01433-4
Choe, J., Hamme , A., & Mon gome y, C. (2019). U.S. SOYBEAN EXPORTS TO CHINA CRUSHED
AMID RISING TRADE TENSIONS.
Cohn, A. S., Mosnie , A., Ha lík, P., Valin, H., He e o, M., Schmid, E., O’Ha e, M., &
Obe s eine , M. (2014). Ca le anching in ensi ica ion in B azil can educe global
g eenhouse gas emissions by spa ing land om de o es a ion. P oceedings o he
Na ional Academy o Sciences, 111(20), 7236–7241.
h ps://doi.o g/10.1073/pnas.1307163111
Co ell, R. S., Blancha d, J. L., Halpe n, B. S., Me ian, M., & F oehlich, H. E. (2020). Global
adop ion o no el aquacul u e eeds could subs an ially educe o age ish demand by
2030. Na u e Food, 1(5), 301–308. h ps://doi.o g/10.1038/s43016-020-0078-x
Co ell, R. S., Fleming, A., Ful on, E. A., Nash, K. L., Wa son, R. A., & Blancha d, J. L. (2018).
Conside ing land–sea in e ac ions and ade-o s o ood and biodi e si y. Global
Change Biology, 24(2), 580–596. h ps://doi.o g/10.1111/gcb.13873
Cowley, C. (2020). Reshu ling in Soybean Ma ke s ollowing Chinese Ta i s. The Fede al
Rese e Bank o Kansas Ci y Economic Re iew.
h ps://doi.o g/10.18651/e / 105n1cowley
Cu is, P. G., Slay, C. M., Ha is, N. L., Tyuka ina, A., & Hansen, M. C. (2018). Classi ying d i e s
o global o es loss. Science, 361(6407), 1108–1111.
h ps://doi.o g/10.1126/science.aau3445
83
De Ma ia, M., Robinson, E., Kangile, J. R., Kadigi, R. M., D eoni, I., Cou o, M., Howai, N., Peci, J.,
& Fiennes, S. (2020). Global Soybean T ade – The Geopoli ics o a Bean. UN
En i onmen P og amme Wo ld Conse a ion Moni o ing Cen e (UNEP-WCMC).
h ps://doi.o g/10.34892/7YN1-K494
De Oli ei a Sil a, R., Ba ioni, L. G., Quei oz Pelleg ino, G., & Mo an, D. (2018). The ole o
ag icul u al in ensi ica ion in B azil’s Na ionally De e mined Con ibu ion on emissions
mi iga ion. Ag icul u al Sys ems, 161, 102–112.
h ps://doi.o g/10.1016/j.agsy.2018.01.003
Elleby, C., Domínguez, I. P., Nielsen, R., Nielsen, M., & Ho , A. (2025). In oducing maximum
sus ainable yield a ge s in ishe ies could enhance global ood secu i y.
Communica ions Ea h & En i onmen , 6(1), 33. h ps://doi.o g/10.1038/s43247-024-
01851-4
FAO (Ed.). (2018). The S a e o Wo ld Fishe ies and Aquacul u e 2018—Mee ing he sus ainable
de elopmen goals.
FAO. (2020). Global Fo es Resou ces Assessmen 2020. FAO ;
h ps://openknowledge. ao.o g/handle/20.500.14283/ca9825en
FAOSTAT. (2025). Fo es P oduc ion and T ade. h ps://www. ao.o g/ aos a /en/#da a/FO
F ezal, C., & Deuss, A. (2025, Feb ua y 11). T ade- ela ed measu es linked o he en i onmen al
sus ainabili y o ag icul u e. OECD. h ps://www.oecd.o g/en/publica ions/ ade-
ela ed-measu es-linked- o- he-en i onmen al-sus ainabili y-o -ag icul u e_eb dca09-
en.h ml
F icko, O., Ha lik, P., Rogelj, J., Klimon , Z., Gus i, M., Johnson, N., Kolp, P., S ubegge , M.,
Valin, H., Amann, M., E molie a, T., Fo sell, N., He e o, M., Heyes, C., Kinde mann, G.,
K ey, V., McCollum, D. L., Obe s eine , M., Pachau i, S., … Riahi, K. (2017). The ma ke
quan i ica ion o he Sha ed Socioeconomic Pa hway 2: A middle-o - he- oad scena io
o he 21s cen u y. Global En i onmen al Change, 42, 251–267.
h ps://doi.o g/10.1016/j.gloen cha.2016.06.004
F oehlich, H. E., Jacobsen, N. S., Essing on, T. E., Cla elle, T., & Halpe n, B. S. (2018). A oiding
he ecological limi s o o age ish o ed aquacul u e. Na u e Sus ainabili y, 1(6), 298–
303. h ps://doi.o g/10.1038/s41893-018-0077-1
F oehlich, H. E., Runge, C. A., Gen y, R. R., Gaines, S. D., & Halpe n, B. S. (2018). Compa a i e
e es ial eed and land use o an aquacul u e-dominan wo ld. P oceedings o he
Na ional Academy o Sciences, 115(20), 5295–5300.
h ps://doi.o g/10.1073/pnas.1801692115
Fulle , M. R., Ganjam, M., Bake , J. S., & Ab , R. C. (2025). Ad ancing o es ca bon p ojec ions
equi es imp o ed con e gence be ween ecological and economic models. Ca bon
Balance and Managemen , 20(1), 2. h ps://doi.o g/10.1186/s13021-024-00290-0
Ful io, F. D., Snäll, T., Lau i, P., Fo sell, N., Mönkkönen, M., Bu gas, D., Bla e , C., Ey indson,
K., Caicoya, A. T., Ve ga echea, M., An ón-Fe nández, C., Klein, J., As up, R.,
Lukka inen, J., Pi zén, S., & P imme , E. (2025). Impac o he EU biodi e si y s a egy
84
o 2030 on he EU wood-based bioeconomy. Global En i onmen al Change, 92,
102986. h ps://doi.o g/10.1016/j.gloen cha.2025.102986
Gepha , J. A., Hen iksson, P. J. G., Pa ke , R. W. R., Shepon, A., Go ospe, K. D., Be gman, K.,
Eshel, G., Golden, C. D., Halpe n, B. S., Ho nbo g, S., Jonell, M., Me ian, M., Mi lin, K.,
New on, R., Tyedme s, P., Zhang, W., Ziegle , F., & T oell, M. (2021). En i onmen al
pe o mance o blue oods. Na u e, 597(7876), 360–365.
h ps://doi.o g/10.1038/s41586-021-03889-2
Guido i, V., F ei as, F. L. M., Spa o ek, G., Pin o, L. F. G., Hamamu a, C., Ca alho, T., &
Ce ignoni, F. (2017). Núme os de alhados do no o Código Flo es al e suas implicações
pa a os PRAs. h ps:// eposi o io.usp.b /i em/002937819
Hadjikakou, M., Bowles, N. I., Geyik, O., Conijn, S. J. G., Mogollón, J. M., Bodi sky, B. L., Mulle ,
A., Weindl, I., Moallemi, E. A., Shaikh, M. A., Dame au, K., Da is, K. F., P is e , S.,
Sp ingmann, M., Cla k, M., Me son, G. S., Röös, E., Bajzelj, B., G aham, N. T., … B yan,
B. A. (2025). Ambi ious ood sys em in e en ions equi ed o mi iga e he isk o
exceeding Ea h’s en i onmen al limi s. One Ea h, 101351.
h ps://doi.o g/10.1016/j.oneea .2025.101351
Halpe n, B. S., F azie , M., Ve s aen, J., Rayne , P.-E., Clawson, G., Blancha d, J. L., Co ell, R.
S., F oehlich, H. E., Gepha , J. A., Jacobsen, N. S., Kuempel, C. D., McIn y e, P. B.,
Me ian, M., Mo an, D., Nash, K. L., Többen, J., & Williams, D. R. (2022). The
en i onmen al oo p in o global ood p oduc ion. Na u e Sus ainabili y, 5(12), 1027–
1039. h ps://doi.o g/10.1038/s41893-022-00965-x
Hende s, S., Pe sson, U. M., & Kas ne , T. (2015). T ading o es s: Land-use change and ca bon
emissions embodied in p oduc ion and expo s o o es - isk commodi ies.
En i onmen al Resea ch Le e s, 10(12), 125012. h ps://doi.o g/10.1088/1748-
9326/10/12/125012
Humpenöde , F., Popp, A., Die ich, J. P., Klein, D., Lo ze-Campen, H., Bonsch, M., Bodi sky, B.
L., Weindl, I., S e ano ic, M., & Mülle , C. (2014). In es iga ing a o es a ion and
bioene gy CCS as clima e change mi iga ion s a egies. En i onmen al Resea ch
Le e s, 9(6), 064029. h ps://doi.o g/10.1088/1748-9326/9/6/064029
IBF-IIASA. (2023). Global Biosphe e Managemen Model (GLOBIOM) Documen a ion 2023—
Ve sion 1.0. In e na ional Ins i u e o Applied Sys ems Analysis.
h ps://pu e.iiasa.ac.a /18996
IFPRI. (2025a, Ap il 8). How ‘ ecip ocal a i s’ ha m ag icul u al ade. IFPRI.
h ps://www.i p i.o g/blog/how- ecip ocal- a i s-ha m-ag icul u al- ade/
IFPRI. (2025b, Ap il 21). U.S.-China ade wa 2.0: Wha a e he implica ions o global oilseed
ma ke s? IFPRI. h ps://www.i p i.o g/blog/u-s-china- ade-wa -2-0-wha -a e- he-
implica ions- o -global-oilseed-ma ke s/
IIASA. (2018). SSP Da abase—Ve sion 2.0.
h ps:// n ca .iiasa.ac.a /SspDb/dsd?Ac ion=h mlpage&page=welcome
IIASA. (2021). ENGAGE Scena io Explo e . h ps://da a.ece.iiasa.ac.a /engage/#/login
85
IIASA. (2025). GLOBIOM- o es documen a ion. Gi Hub.
h ps://gi hub.com/iiasa/GLOBIOM_ o es /blob/main/GLOBIOM_ o es _documen a i
on.pd
IPCC. (2019). Ha es ed wood p oduc s—2019 Re inemen o he 2006 IPCC Guidelines o
Na ional G eenhouse Gas In en o ies (Vol. 4). h ps://www.ipcc-
nggip.iges.o .jp/public/2019 /pd /4_Volume4/19R_V4_Ch12_Ha es edWoodP oduc
s.pd
Johns on, C. M. T., & Radelo , V. C. (2019). Global mi iga ion po en ial o ca bon s o ed in
ha es ed wood p oduc s. P oceedings o he Na ional Academy o Sciences, 116(29),
14526–14531. h ps://doi.o g/10.1073/pnas.1904231116
Lau i, P., Fo sell, N., Di Ful io, F., Snäll, T., & Ha lik, P. (2021). Ma e ial subs i u ion be ween
coni e ous, non-coni e ous and ecycled biomass – Impac s on o es indus y aw
ma e ial use and egional compe i i eness. Fo es Policy and Economics, 132, 102588.
h ps://doi.o g/10.1016/j. o pol.2021.102588
Leclè e, D., Obe s eine , M., Ba e , M., Bu cha , S. H. M., Chaudha y, A., De Palma, A.,
DeCle ck, F. A. J., Di Ma co, M., Doelman, J. C., Dü aue , M., F eeman, R., Ha oo , M.,
Hasegawa, T., Hellweg, S., Hilbe s, J. P., Hill, S. L. L., Humpenöde , F., Jennings, N.,
K isz in, T., … Young, L. (2020). Bending he cu e o e es ial biodi e si y needs an
in eg a ed s a egy. Na u e, 585(7826), 551–556. h ps://doi.o g/10.1038/s41586-020-
2705-y
Lesi , M., Schepaschenko, D., Buchho n, M., See, L., Dü aue , M., Geo gie a, I., Jung, M.,
Ho hansl, F., Schulze, K., Bilous, A., Blyshchyk, V., Mukho o a, L., B enes, C. L. M.,
K i oboko , L., N ie, S., Tsog , K., Pie sch, S. A., Tikhono a, E., Kim, M., … F i z, S.
(2022). Global o es managemen da a o 2015 a a 100 m esolu ion. Scien i ic Da a,
9(1), 199. h ps://doi.o g/10.1038/s41597-022-01332-3
Mish a, A., Humpenöde , F., Chu kina, G., Reye , C. P. O., Beie , F., Bodi sky, B. L.,
Schellnhube , H. J., Lo ze-Campen, H., & Popp, A. (2022). Land use change and ca bon
emissions o a ans o ma ion o imbe ci ies. Na u e Communica ions, 13(1), 4889.
h ps://doi.o g/10.1038/s41467-022-32244-w
Mish a, A., Humpenöde , F., Die ich, J. P., Bodi sky, B. L., Sohngen, B., P. O. Reye , C., Lo ze-
Campen, H., & Popp, A. (2021). Es ima ing global land sys em impac s o imbe
plan a ions using MAgPIE 4.3.5. Geoscien i ic Model De elopmen , 14(10), 6467–6494.
h ps://doi.o g/10.5194/gmd-14-6467-2021
Naylo , R. L., Kisho e, A., Sumaila, U. R., Issi u, I., Hun e , B. P., Bel on, B., Bush, S. R., Cao, L.,
Gelcich, S., Gepha , J. A., Golden, C. D., Jonell, M., Koehn, J. Z., Li le, D. C., Thils ed, S.
H., Tigchelaa , M., & C ona, B. (2021). Blue ood demand ac oss geog aphic and
empo al scales. Na u e Communica ions, 12(1), 5413.
h ps://doi.o g/10.1038/s41467-021-25516-4
OECD/FAO. (2024, July 2). OECD-FAO Ag icul u al Ou look 2024-2033.
h ps://www.oecd.o g/en/publica ions/oecd- ao-ag icul u al-ou look-2024-
2033_4c5d2c b-en/ ull- epo .h ml
86
Oli ei a, U., & Pacheco, A. (2024). Deli e able 2.4: Biodi e si y impac da abase including
cha ac e iza ion ac o s and documen a ion eady o Module C. Zenodo.
h ps://doi.o g/10.5281/zenodo.13640620
Oli ei a, U., Soa es-Filho, B. S., San os, A. J., Paglia, A. P., B esco i , A. D., de Ca alho, C. J. B.,
Sil a, D. P., Rezende, D. T., Lei e, F. S. F., Ba is a, J. A. N., Ba bosa, J. P. P. P., S ehmann,
J. R., Asche , J. S., Vasconcelos, M. F., Ma co, P. D., Löwenbe g-Ne o, P., & Fe o, V. G.
(2019). Modelling Highly Biodi e se A eas in B azil. Scien i ic Repo s, 9(1), 6355.
h ps://doi.o g/10.1038/s41598-019-42881-9
Pend ill, F., Pe sson, U. M., Goda , J., & Kas ne , T. (2019). De o es a ion displaced: T ade in
o es - isk commodi ies and he p ospec s o a global o es ansi ion. En i onmen al
Resea ch Le e s, 14(5), 055003. h ps://doi.o g/10.1088/1748-9326/ab0d41
Pend ill, F., Pe sson, U. M., Goda , J., Kas ne , T., Mo an, D., Schmid , S., & Wood, R. (2019).
Ag icul u al and o es y ade d i es la ge sha e o opical de o es a ion emissions.
Global En i onmen al Change, 56, 1–10.
h ps://doi.o g/10.1016/j.gloen cha.2019.03.002
Pe ei a, H., Ma ins, I., Rosa, I., Kim, H., & Leadley, P. (2024). Global ends and scena ios o
e es ial biodi e si y and ecosys em se ices om 1900 o 2050.
h ps://doi.o g/10.1126/science.adn3441
Popp, A., Cal in, K., Fujimo i, S., Ha lik, P., Humpenöde , F., S eh es , E., Bodi sky, B. L.,
Die ich, J. P., Doelmann, J. C., Gus i, M., Hasegawa, T., Kyle, P., Obe s eine , M.,
Tabeau, A., Takahashi, K., Valin, H., Waldho , S., Weindl, I., Wise, M., … Vuu en, D. P.
V. (2017). Land-use u u es in he sha ed socio-economic pa hways. Global
En i onmen al Change, 42, 331–345. h ps://doi.o g/10.1016/j.gloen cha.2016.10.002
p o ec ed plane . (2025). P o ec ed A eas (WDPA). P o ec ed Plane .
h ps://www.p o ec edplane .ne /en/ hema ic-a eas/wdpa
Ri ie e, M., Cau la, S., & Delaco e, P. (2020). E ol ing In eg a ed Models F om Na owe
Economic Tools: The Example o Fo es Sec o Models. En i onmen al Modeling &
Assessmen , 25(4), 453–469. h ps://doi.o g/10.1007/s10666-020-09706-w
Robe s, S., Jacque , J., Majlu , P., & Hayek, M. N. (2024). Feeding global aquacul u e. Science
Ad ances, 10(42), eadn9698. h ps://doi.o g/10.1126/sciad .adn9698
Salles, J.-M., Teilla d, F., Tichi , M., & Zanella, M. (2017). Land spa ing e sus land sha ing: An
economis ’s pe spec i e. Regional En i onmen al Change, 17(5), 1455–1465.
h ps://doi.o g/10.1007/s10113-017-1142-4
Schilling-Vaca lo , A., & Lenschow, A. (n.d.). B inging he S a e back in: Explo ing new public
en i onmen al policy app oaches o go e ning he B azil-Eu ope soy chain.
Schippe , A. M., Hilbe s, J. P., Meije , J. R., An ão, L. H., Bení ez-López, A., de Jonge, M. M. J.,
Leemans, L. H., Schepe , E., Alkemade, R., Doelman, J. C., Mylius, S., S eh es , E., an
Vuu en, D. P., an Zeis , W.-J., & Huijb eg s, M. A. J. (2020). P ojec ing e es ial
biodi e si y in ac ness wi h GLOBIO 4. Global Change Biology, 26(2), 760–771.
h ps://doi.o g/10.1111/gcb.14848
87
Schul e, M., Lau i, P., & Di Ful io, F. (2025). Global o es ca bon leakage and subs i u ion
e ec po en ials: The case o he Swedish o es sec o . Ca bon Balance and
Managemen (in Re iew).
Sla e , M., & James, P. (2023). Low ophic species in aquacul u e—G ow h and esea ch
challenges. Jou nal o he Wo ld Aquacul u e Socie y, 54(1), 4–6.
h ps://doi.o g/10.1111/jwas.12944
So e oni, A. C., Mosnie , A., Ca alho, A. X. Y., Câma a, G., Obe s eine , M., And ade, P. R.,
Souza, R. C., B ock, R., Pi ke , J., K axne , F., Ha lík, P., Kapos, V., Zu E mgassen, E. K. H.
J., Valin, H., & Ramos, F. M. (2018). Fu u e en i onmen al and ag icul u al impac s o
B azil’s Fo es Code. En i onmen al Resea ch Le e s, 13(7), 074021.
h ps://doi.o g/10.1088/1748-9326/aaccbb
So i o , M., Aze edo-Ramos, C., Ra is, L., & Be ning, L. (2022). Policy op ions o egula e
imbe and ag icul u al supply-chains o legali y and sus ainabili y: The case o he EU
and B azil. Fo es Policy and Economics, 144, 102818.
h ps://doi.o g/10.1016/j. o pol.2022.102818
Spillias, S., Blancha d, J. L., Cas o-Cadenas, M. D., Co ell, R. S., Deppe man, A., F ank, S.,
Lecle e, D., T an, N., & Ha lik, P. (2025). In eg a ing Blue Foods in o Fu u e Es ima es
o Land-Use Change (manusc ip in p epa a ion).
S eh es e al. (2019). Key de e minan s o global land-use p ojec ions.
h ps://www.na u e.com/a icles/s41467-019-09945-w
Tacon, A. G. J., & Me ian, M. (2015). Feed Ma e s: Sa is ying he Feed Demand o
Aquacul u e. Re iews in Fishe ies Science & Aquacul u e, 23(1), 1–10.
h ps://doi.o g/10.1080/23308249.2014.987209
Uni ed Na ions. (2025). UN Com ade. h ps://com adeplus.un.o g/
Valen i, W. C., & Balles e , E. L. (2024). In eg a ed Aquacul u e and Monocul u e o Low-
T ophic Species. Fishes, 9(11), 450. h ps://doi.o g/10.3390/ ishes9110450
Ve ones, F., Hellweg, S., An ón, A., Aze edo, L. B., Chaudha y, A., Cosme, N., Cucu achi, S., de
Baan, L., Dong, Y., Fan ke, P., Gols eijn, L., Hauschild, M., Heijungs, R., Jollie , O.,
Ju aske, R., La sen, H., Lau en , A., Mu el, C. L., Ma gni, M., … Huijb eg s, M. A. J.
(2020). LC-IMPACT: A egionalized li e cycle damage assessmen me hod. Jou nal o
Indus ial Ecology, 24(6), 1201–1219. h ps://doi.o g/10.1111/jiec.13018
Wo ld Economic Fo um. (2018, May 18). B azil’s o es s could all ic im o he US-China ade
wa . Wo ld Economic Fo um. h ps://www.we o um.o g/s o ies/2018/05/b azil-
o es s-could-be-casual ies-in-u-s-china- ade-wa /
Wo ld Economic Fo um. (2025, May 15). How B azil and China can deli e on sus ainabili y
goals. Wo ld Economic Fo um. h ps://www.we o um.o g/s o ies/2025/05/b azil-and-
china-can- ans o m- he-sus ainabili y-o -soy-supply-chains/
Wu, W., Hasegawa, T., Ohashi, H., Hanasaki, N., Liu, J., Ma sui, T., Fujimo i, S., Masui, T., &
Takahashi, K. (2019). Global ad anced bioene gy po en ial unde en i onmen al
p o ec ion policies and socie al ans o ma ion measu es. GCB Bioene gy, 11(9), 1041–
1055. h ps://doi.o g/10.1111/gcbb.12614
88
Xiao, X., Agus i, S., Lin, F., Li, K., Pan, Y., Yu, Y., Zheng, Y., Wu, J., & Dua e, C. M. (2017).
Nu ien emo al om Chinese coas al wa e s by la ge-scale seaweed aquacul u e.
Scien i ic Repo s, 7(1), 46613. h ps://doi.o g/10.1038/s ep46613
Zhao, J., Wei, X., & Li, L. (2022). The po en ial o s o ing ca bon by ha es ed wood p oduc s.
F on ie s in Fo es s and Global Change, 5. h ps://doi.o g/10.3389/ gc.2022.1055410
Ziege , R. F., & So i o , M. (2024). Regula o y poli ics and hyb id go e nance: The case o
B azil’s Amazon Soy Mo a o ium. Global En i onmen al Change, 88, 102916.
h ps://doi.o g/10.1016/j.gloen cha.2024.102916
Zilli, M., Sca abello, M., So e oni, A. C., Valin, H., Mosnie , A., Leclè e, D., Ha lík, P., K axne ,
F., Lopes, M. A., & Ramos, F. M. (2020). The impac o clima e change on B azil’s
ag icul u e. Science o The To al En i onmen , 740, 139384.
h ps://doi.o g/10.1016/j.sci o en .2020.139384
APPENDIX
S akeholde wo kshop o soy supply chain scena ios
To suppo he model and scena io wo k, a wo-hou online s akeholde wo kshop ook place
on May 14 h 2025.
Po en ial pa icipan s we e iden i ied h ough he CLEVER s akeholde mapping, and 31
pa icipan s esponded posi i ely o he in i a ion, o which 23 pa icipa ed. O he 23
pa icipa ing s akeholde s, 11 we e om NGOs, 6 om he public sec o (including bo h na ional
go e nmen and in e na ional o ganiza ions), and 6 om he p i a e sec o . The majo i y o
pa icipa ing s akeholde s we e based in B azil and Eu ope, and 2 we e based in No he n
Ame ica.
The goal o he wo kshop was o ga he inpu om s akeholde s on ele an scena ios and
modelling ea u es o he soy sec o and i s impac on biodi e si y, wi h a speci ic ocus on
B azil-EU soy supply chains. The wo kshop consis ed o : a) a 10 min welcome and in oduc ion
session, b) a 20min p esen a ion om IIASA on model imp o emen s, po en ial scena io op ions
and p elimina y esul s, ollowed by a Q&A, c) a i s b eakou session whe e expe s we e asked
o sha e hei ision o he soy sec o by 2050 (4 b eakou g oups o 10min ollowed by a 10min
epo ing back in plena y, wi h no e aking on he whi eboa d), d) a second b eakou session
whe e expe s we e asked o iden i y hei p e e ed scena io op ions and discuss he likely
impac o he selec ed policies/s ylized in e en ions on soy expo s and biodi e si y (4 b eakou
g oups o 10min ollowed by a 10min epo ing back in plena y, wi h no e aking on he
whi eboa d), and e) a 5min closing session. All pa icipan s we e p o ided wi h he p esen a ion
and he whi eboa d, as well as a link o a o m o any addi ional eedback hey wish o p o ide.
Figu e 44. Linkedin pos pos ed a e he wo kshop, which includes a sc eensho o he whi e boa d, o IIASA
p esen a ion and o he zoom pa icipan s.
96
Figu e 48 - Compa ison o obse ed (IBGE PAM) s simula ed (GLOBIOM) ends in a ious aspec s (ha es ed
a ea, p oduc ion olume, yield and sha e o soy-co n double c opping in o al ha es ed a ea) o soy p oduc ion
sys ems o e 2000-2020 a he biome le el. Fo each aspec and biome, alues a e ime ho izon- and sou ce-
speci ic (i.e., obse ed s simula ed) alues no malized by he alue o yea 2000 in he obse a ion da ase .
Model imp o emen s o o es supply chains
De ails on he modelling amewo k o o es supply chains can be ound in he deli e able D6.2
while addi ional model imp o emen s suppo ing his deli e able (i.e. inclusion o o es age
s uc u e, HWP and BECCS accoun ing, na u al land di ision) a e de ailed below.
Age-class dynamics and o es managemen
Fo es age-class dynamics in GLOBIOM wo ks simila ly han in la ge-scale a ea-based ma ix
models such as he Eu opean Fo es y Dynamics Model (EFDM) (Packalen e al. 2023) o he
Eu opean Fo es In o ma ion Scena io Model (EFISCEN) (Schelhaas e al. 2007). The main
di e ence be ween GLOBIOM and hese models is ha ha es decisions and age-class dynamics
a e based on economic op imiza ion ins ead o s a is ical ansi ion p obabili ies.
In GLOBIOM, he choice o e di e en managemen sys ems and age-classes is made sepa a ely
o each g id. Each g id consis s o mul iple e en-aged s ands which a e di ided o di e en age-
coho s. Biomass g ow h happens h ough age-class dynamics whe e a e each pe iod he s and
a ea is mo ed o he nex age-coho . The biomass in each age-coho ollows S-shaped g ow h
cu es which a e de e mined by he Chapman-Richa ds biomass g ow h model, as in
(Humpenöde e al., 2014) and (Mish a e al., 2021). I he s and a ea is ha es ed o damaged
by na u al mo ali y, hen i is mo ed in he i s age-coho . Since he size o he s and is no
de ined explici ly, he age-class dynamics can be in e p e ed ei he as e en-o une en aged
managemen s.
Figu e 49 displays an example o age-class dynamics in GLOBIOM. In 2020, he ca bon s ock o
o es s is calib a ed o (FAO, 2020) da a and downscaled o g id le el by u ilizing G4M da a and
global age-class da abase (Besna d e al., 2021) . A e 2020, he ca bon s ock is de e mined
endogenously by he age-class dynamics. The ca bon s ock o o es s is highe in 2100 han in
2020, because he es ima ed biomass g ow h exceeds he es ima ed biomass emo als om he
o es s (ha es s, mo ali y) du ing his pe iod.
98
Figu e 49. Global ca bon s ock o o es s di ided by age-classes in he BASE scena io (excluding p ima y o es s)
In GLOBIOM, o es s a e di ided in o h ee classes: 1) Plan a ion o es s, 2) Semina u al o es s
and 3) Na u al o es s. Wi hin each o es class he e a e di e en managemen sys ems as
showed in Table 4. The cu en o mula ion o he model includes 9 o es managemen sys ems.
Table 4. Fo es managemen sys ems in GLOBIOM
Managemen
Desc ip ion
Fo es ype
Managemen in ensi y
Calib a ion
Cu C_H
Coni e ous plan a ion
o es
Plan a ion o es
3PGmix sho o a ion
FRA2020+Lesi 2022
plan a ion o es a ea
Cu NC_H
Non-coni e ous
plan a ion o es
Plan a ion o es
3PGmix sho o a ion
FRA2020+Lesi 2022
plan a ion o es a ea
Cu C
Coni e ous
p oduc ion o es
Semi-na u al o es
G4M EU 75-100%
ROW 50-100 %
FRA2020+Lesi 2022
p oduc ion o es
a ea
Cu NC
Non-coni e ous
p oduc ion o es
Semi-na u al o es
G4M EU 75-100%
ROW 50-100 %
FRA2020+Lesi 2022
p oduc ion o es
a ea
Cu C_M
Coni e ous
mul i unc ional o es
Semi-na u al o es
G4M EU 50-75%
ROW 25-50 %
FRA2020+Lesi 2022
p oduc ion o es
a ea
Cu NC_M
Non-coni e ous
mul i unc ional o es
Semi-na u al o es
G4M EU 50-75%
ROW 25-50 %
FRA2020+Lesi 2022
p oduc ion o es
a ea
Cu C_L
Coni e ous close- o-
na u e o es
Na u al o es
G4M EU 0-50%
ROW 0-25 %
WDPA ca ego ies
IV-VI
Cu NC_L
Non-coni e ous
close- o- na u e
o es
Na u al o es
G4M EU 0-50%
ROW 0-25 %
WDPA ca ego ies
IV-VI
Cu S
S ic ly p o ec ed
seconda y o es
Na u al o es
0%
WDPA ca ego ies
I-III
Cu 0
Unp o ec ed
seconda y o es
Na u al o es
0%
Residual o es a ea
0
5
10
15
20
25
30
35
40
0-10
10_20
20-30
30-40
40-50
50-60
60-70
70-80
80-90
90-100
100-110
110-120
120-130
130-140
140-150
150+
PgC
BASE 2020
0
5
10
15
20
25
30
35
40
0-10
10_20
20-30
30-40
40-50
50-60
60-70
70-80
80-90
90-100
100-110
110-120
120-130
130-140
140-150
150+
PgC
BASE 2100
99
P iFo
P ima y o es
Na u al o es
0%
FRA2020+Lesi 2022
p ima y o es a ea,
EU Saba ini map
No e: 1) Managemen in ensi y=% sha e o G4M maximum sus ainable ha es po en ial based on NNP maps, local
ee species and o a ion ime ha maximizes inc emen gi en han ca bon s ock s ays a he cu en le el.
2) The model can choose be ween coni e ous and non-coni e ous managemen op ion in he limi s o ee species
dis ibu ion based on FRA2020 da a (FAO, 2020). In he EU, he model use addi ional g id le el da a on ee species
om (B us e al., 2012).
The ansi ion om na u al o semi-na u al o es s o plan ed o es happens h ough in ensi e
managemen , i.e., emo ing old ees by ha es ing and plan ing new ones. The ansi ion om
plan ed o es s o na u al o semi-na u al o es s happens h ough ex ensi e managemen , i.e.,
emo ing old ees by ha es ing o h ough na u al mo ali y and wai ing o new ees o
appea by na u al egene a ion.
The spa ially explici biophysical da a o each g id is based on he G4M o es managemen
simula ion model (Kinde man e al. 2006,2008). G4M p o ides GLOBIOM es ima es on ini ial
biomass s ock, g owing s ock and sus ainable le el o ha es s (=inc emen ). Using he G4M da a
and global age-class da abase (Besna d e al. 2021), he model gene a es g id le el biomass
g ow h cu es, i.e., biomass and g owing s ocks o di e en s ands wi hin each g id.
Wi hin each g id he model can ha e di e en managemen sys ems wi h di e en age-class
dynamics, mo ali y and ha es in ensi y. The ini ial a eas o di e en managemen sys ems
a e calib a ed o he FRA (2020) da a, he Wo ld Da abase o p o ec a eas da a (p o ec ed
plane , 2025) and he Global Fo es Managemen Map (Lesi e al., 2022). Mo eo e , o he
his o ical pe iods 2000-2020 he managemen a eas a e ma ched o FAOSTAT coun y-le el
ha es olumes da a (FAOSTAT, 2025).
A e 2020, he age-class dynamics de elops endogenously based on pe iodic ha es s olumes,
g ow h cu es, mo ali y and o es a ea changes (de o es a ion/a o es a ion). To main ain he
sus ainabili y o ha es olumes o e ime in he ecu si e dynamics, he model includes wo
addi ional cons ain s. Fi s , o each g id cell, he amoun o ha es ed biomass canno exceed
biomass g ow h. Second, i is no allowed o ha es age-classes ha a e younge han he
op imal o a ion ime. Op imal o a ion imes o na u al/semina u al o es s a y be ween 30-
80 yea s depending on he managemen sys em and clima e zone while o plan a ion o es s
be ween 5-30 yea s.
Finally, he ini ial calib a ion o he model o he his o ical pe iod 2000-2020 has been imp o ed
o be e ma ch be e ha es olumes da a (FAOSTAT, 2025), o es biomass s ocks and ca bon
emissions da a (FAO, 2020) and o es a eas and land-use changes da a (FAO, 2020).
HWP and BECCS ca bon accoun ing
Ha es ed wood p oduc s (HWP) ca bon accoun ing ollows he IPCC p oduc ion app oach (PA),
which is he mos commonly used me hodology o calcula e HWP ca bon s o age (IPCC, 2019).
The ini ial HWP ca bon pool is calcula ed by using he p oduc ion and ade da a o HWP om
FAOSTAT da abase om 1961 o 2020 (FAOSTAT, 2025). Based on his me hod he global HWP
100
pool was a sink o abou 300 M CO2/y in 2020, which is compa able o global HWP pool
es ima es by (Johns on & Radelo , 2019). The HWP ca bon s o age and sink a e conside able
smalle han o he o es ca bon s o age and sink, ypically HWP ca bon s o age is abou 5% o
o es abo e g ound ca bon s o age and HWP ca bon sink abou 10% o o es ca bon sink (Zhao
e al., 2022). The e a e wo easons o his. Fi s , abou 75% o ee biomass is los o ha es
and o es indus y esidues when woody biomass is mo ed away om o es s and con e ed
o HWP. Second, he li e ime HPW (0-100 yea s depending on he p oduc ) is ypically sho e
han he li e ime o ees (10-500 yea s depending on he managemen ).
In addi ion o HWP ca bon s o age, we also calcula e bioene gy wi h ca bon cap u e and s o age
(BECCS) ca bon s o age. The di e ence be ween hem is ha a hal -li e HWP a y om 0 o 35
yea s while BECCS has an in ini e hal -li e, i.e., HWP a e empo a y ca bon s o ages while BECCS
is a pe manen ca bon s o age. The sha e o bioene gy wi h BECCS is p o ided by he MESSAGE
model and inc eases om 5% in 2030 o 75% in 2100 (IIASA, 2018). The BECCS ca bon cap u e
a e ypically a ies be ween 25-75% depending on he inal ene gy ca ie . We assume ha
BECCS ca bon cap u e a e is 63%, which is an a e age alue es ima ed om he MESSAGE
model ou come.
De o es a ion and a o es a ion
In his o ical pe iods 2000-2020, de o es a ion and a o es a ion a eas a e based on (FAO, 2020)
coun y le el da a, which is alloca ed in he g id le el acco ding o GLOBIOM land-use dynamics.
The land-use dynamics is based on land-use change cos s, land sui abili y maps, p oduc ion
po en ial maps and demand o land on ood, eed and ibe p oduc ion (IBF-IIASA, 2023).
A e 2020, de o es a ion a ea is endogenous and depends on ood and eed demand as well as
ca bon ax on de o es a ion emissions. Because ca bon p ices a e ela i ely high in RCP1p9
scena io, he e is e y li le de o es a ion a e 2020 (Figu e 50). Rema k ha con e sion o
na u al o es s o plan a ions is excluded in he model, i.e., he mo i e o de o es a ion in he
model is ood and eed demand ins ead o imbe demand o some o he eason (FAO, 2020).
A e 2020, a o es a ion a ea is based on G4M SSP2 RCP1p9 coun y le el scena io da a, which
is alloca ed in he g id le el acco ding o GLOBIOM land-use dynamics. A o es a ion compe es
wi h ene gy c ops plan a ions, o es plan a ions and ood p oduc ion on he same ma ginal
land a eas. The di e ence be ween a o es a ion and o es plan a ions is ha a o es ed a eas
a e managed o ca bon seques a ion and na u e conse a ion while o es plan a ions a ea
managed o p oduc ion use. The e a e also some di e ences in a o es a ion and plan a ions
land sui abili y maps, i.e., some a eas a e sui able o a o es a ion bu no o plan a ions (low
p oduc i i y a eas, deep slopes e c.).
101
Figu e 50. Global de o es a ion and a o es a ion
O he na u al land di ision o abandoned land and na u al land
In GLOBIOM, o he na u al land is di ided in wo land-co e classes: 1) Abandoned land
(AbdLnd), and 2) Na u al land (Na Lnd). AbdLnd is abandoned ag icul u al land, which ypically
has low biodi e si y alue and deg aded ca bon s ock. Na Lnd is old-g ow h na u al ege a ion
land such as sa annas, p ai ies and open o es s (no classi ied as o es s), which ha e high
biodi e si y alue and ca bon s ock. AbdLnd is allowed o be con e ed o o he land-use
acco ding o GLOBIOM usual land-use change cons ain s (land con e sion cos s + maximum
a ailable land) while Na Lnd con e sion o o he land-use is limi ed by applying highe land
con e sion cos s o Na Lnd han o AbdLnd. Highe land con e sion cos s elimina e mos o
Na Lnd con e sion o o he land uses.
Since no good map o his o ically abandoned lands is a ailable, we le he model decide he le el
o a ailable AbdLnd o each pe iod. Based on his me hod, he AbdLnd a ea inc ease om ze o
in 2000 o 270 Mha in 2040, and hen s a s dec easing depending on he scena io (Figu e 51).
In gene al, AbdLnd dec eases o e ime, because i is con e ed o plan a ions, o es s o
ag icul u al land. On he o he hand, some ag icul u al land a ea is abandoned due o changes
in demand and/o p oduc i i y, which inc eases AbdLnd a ea o e ime.
0
100
200
300
400
500
600
700
800
900
2000 2020 2040 2060 2080
2100
Mha
D Lnd
0
100
200
300
400
500
600
700
800
900
2000 2020 2040 2060 2080
2100
Mha
A Lnd
102
Figu e 51. Global abandoned land (AbdLnd) de elopmen ela i e na u al land (Na Lnd) in he baseline scena io
Wood-based p oduc s ade and egional compe i i eness
The GLOBIOM model includes 22 aded wood-based p oduc s, which di e in uni s and le el o
alue-addi ion (i.e., p ima y, semi inished, inal). Rema k ha ene gy c ops a e no included in
he aded p oduc s in he model. Reason is ha ene gy c ops a e low alue-add p oduc s, which
a e usually no anspo ed longe dis ances o aded be ween he coun ies.
Fo simplici y, we do no conside he e indi idual wood-based p oduc s bila e al ade lows,
bu only o al ne expo s (i.e. di e ence be ween expo olume and impo olume agg ega ed
o e all p oduc s). The olume o o al ne expo s e lec s egional woody biomass esou ces,
p oduc ion cos s and o es indus y de elopmen s a e, and can be used as an indica o o
egional economic compe i i eness.
Agg ega ion o e di e en wood-based p oduc s equi es ha hey a e con e ed in
compa able uni s. The e a e di e en me hods o con e wood-based p oduc s o compa able
uni s (Lau i e al. 2021). He e we use he e so called oundwood equi alen (RWeq) uni s, which
measu e wood-based p oduc s acco ding o he amoun o p ima y biomass needed o hei
p oduc ion. This me hod is s anda d in he o es sec o analysis and ecen ly applied also in he
ag icul u al sec o analysis (Zhao e al. 2025).
0
500
1000
1500
2000
2500
3000
3500
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Mha
Na Lnd AbdLnd
103
Model imp o emen s o aquacul u e and aqua eed supply chains
P ojec ing he capaci y o un ed aquacul u e beyond 2020
As compa ed o he desc ip ion o he BAU scena io in D6.2, and in acco dance wi h he eedback
om he s akeholde du ing he wo kshop, o all aquacul u e and aqua eed scena ios in his
deli e able (excep o UNFED20), we assume ha he capaci y o un ed aquacul u e can
inc ease beyond 2020 le els.
To cons ain he pos -2020 capaci y o un ed aquacul u e, we ex apola ed a coun y- and
p oduc -le el his o ical ends using a Gene alized Addi i e Models. Fu u e ends in capaci y
a e p o ided in Figu e 52.
Figu e 52 - P ojec ions o he capaci y o un ed aquacul u e by egion and p oduc
Sensi i i y analysis o he c op composi ion o c op aqua eed equi emen s
Fo all aqua eed and aquacul u e scena ios excep o he AFCOMPOCROPMIX scena io, we
assumed he p opo ion o a ious c ops (be ween soya, co n, ape and whea ) in o al c op
aqua eed demand o be coun y-speci ic (based on F ohlich e al 2018), bu in a ian ac oss ime
ho izons and scena ios.
As his migh no necessa ily be a alid assump ion, and migh a ec he po en ial o impac o
c op aqua eed equi emen s on land use and biodi e si y, we es ed he impac o an al e na i e
assump ion in a simple scena io. In his al e na i e scena io (AFCOMPOCROPMIX), we ocused
on he dominan ed aquacul u e p oduc ion sys em ( ed aquacul u e in China), o which we
assumed ha 50% o he c ude p o ein con en om co n (which has ela i ely lowe c ude
p o ein con en ) and soya (which is impo ed) eed equi emen s pe uni o aquacul u e ou pu
a e eplaced by a simila c ude p o ein con en om whea . To do so, pa ame e s o co n and
soya aqua eed equi emen s whe e hal ed, while ha o whea was inc eased by he amoun
104
necessa y o compensa e his loss in e ms o c ude p o ein in ake (while accoun ing o
di e ences ac oss c ops in c ude p o ein con en and appa en diges ibili y, see Table 5).
Table 5 - Nu i ional pa ame e s o indi idual c ops as aqua eed. Sou ce: INRAE-CIRAD-AFZ Feed ables
(h ps://www. eed ables.com/, accessed July 2024).
CROP
C ude p o ein con en (g am
c ude p o ein pe g am o
eshma e )
Appa en diges ibili y
( ac ion)
Co n
0.216
0.92
Soya
0.404
0.912
Whea
0.404
0.95
