We ne , Cons anze; Luch , Wol gang; Kammann, Claudia; B aun, Johanna
A icle — Published Ve sion
Land-neu al nega i e emissions h ough biocha -based
e iliza ion—assessing global po en ials unde a ied
managemen and py olysis condi ions
Mi iga ion and Adap a ion S a egies o Global Change
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: We ne , Cons anze; Luch , Wol gang; Kammann, Claudia; B aun, Johanna (2024) :
Land-neu al nega i e emissions h ough biocha -based e iliza ion—assessing global po en ials
unde a ied managemen and py olysis condi ions, Mi iga ion and Adap a ion S a egies o Global
Change, ISSN 1573-1596, Sp inge Ne he lands, Do d ech , Vol. 29, Iss. 5,
h ps://doi.o g/10.1007/s11027-024-10130-8
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315329
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
Mi ig Adap S a eg Glob Change (2024) 29:34
h ps://doi.o g/10.1007/s11027-024-10130-8
1 3
ORIGINAL ARTICLE
Land‑neu al nega i e emissions h oughbiocha ‑based
e iliza ion—assessing global po en ials unde a ied
managemen andpy olysis condi ions
Cons anzeWe ne 1,2,3 · Wol gangLuch 1,2,3· ClaudiaKammann4· JohannaB aun1
Recei ed: 30 Sep embe 2022 / Accep ed: 19 Ma ch 2024 / Published online: 5 Ap il 2024
© The Au ho (s) 2024
Abs ac
Clima e s abiliza ion is c ucial o es abilizing he Ea h sys em bu should no unde mine
biosphe e in eg i y, a second pilla o Ea h sys em unc ioning. This is o pa icula con-
ce n i i is o be achie ed h ough biomass-based nega i e emission (NE) echnologies ha
compe e o land wi h ood p oduc ion and ecosys em p o ec ion. We assess he NE con-
ibu ion o land- and calo ie-neu al py ogenic ca bon cap u e and s o age (LCN-PyCCS)
acili a ed by biocha -based e iliza ion, which seques e s ca bon and educes land demand
by inc easing c op yields. Applying he global biosphe e model LPJmL wi h an enhanced
ep esen a ion o as -g owing species o PyCCS eeds ock p oduc ion, we calcula ed a
land-neu al global NE po en ial o 0.20–1.10 G CO2 yea −1 assuming 74% o he biocha
ca bon emaining in he soil a e 100 yea s ( o + 10% yield inc ease; no po en ial o +
5%; 0.61–1.88 G CO2 yea −1 o + 15%). The po en ial is p ima ily d i en by he achie -
able yield inc ease and he managemen in ensi y o he biomass p oducing sys ems. NE
p oduc ion is es ima ed o be enhanced by + 200–270% i managemen in ensi y inc eases
om a ma ginal o a mode a e le el. Fu he mo e, ou esul s show sensi i i y o p ocess-
speci ic biocha yields and ca bon con en s, p oducing a di e ence o + 40–75% be ween
conse a i e assump ions and an op imized se ing. Despi e hese challenges o making
wo ld-wide assump ions on LCN-PyCCS sys ems in modeling, ou indings poin o dis-
c epancies be ween he la ge NE olumes calcula ed in demand-d i en and economically
op imized mi iga ion scena ios and he po en ials om analyses ocusing on supply-d i en
app oaches ha mee en i onmen al and socioeconomic p econdi ions as deli e ed by
LCN-PyCCS.
* Cons anze We ne
cons anze.we ne @pik-po sdam.de
1 Po sdam Ins i u e o Clima e Impac Resea ch, Membe o heLeibniz Associa ion,
Teleg aphenbe g, D-14473Po sdam, Ge many
2 Depa men o Geog aphy, Humbold -Uni e si ä zu Be lin, Un e den Linden 6, D-10099Be lin,
Ge many
3 In eg a i e Resea ch Ins i u e onT ans o ma ions o Human-En i onmen Sys ems, Un e den
Linden 6, D-10099Be lin, Ge many
4 Depa men o Applied Ecology, Hochschule Geisenheim Uni e si y, Von-Lade S . 1,
D-65366Geisenheim, Ge many
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 2 o 28
Keywo ds Ca bon dioxide emo al· Nega i e emissions· Biocha · Py olysis· PyCCS
1 In oduc ion
Nega i e emissions (NE; see Table1 o a lis o abb e ia ions) pose a signi ican and com-
plex challenge o science and policy sea ching o easible pa hways o achie e he clima e
a ge s o he Pa is Ag eemen . In addi ion o deep emission educ ions, NEs a e being
conside ed o o se esidual ha d- o-aba e emissions, bu also o compensa e o delays
in s ingen deca boniza ion. Ye , while especially biomass-based nega i e emissions ech-
nologies (NETs) like bioene gy wi h ca bon cap u e and s o age (BECCS) a e conside ed
easible in economic op imiza ion, land-based op ions equi e as a eas, he eby compe -
ing wi h ood p oduc ion and ecosys em p o ec ion (Boysen e al. 2017; Heck e al. 2018;
Humpenöde e al. 2018). As an al e na i e o land-demanding BECCS, we he e assess ea-
sible NE con ibu ions o mo e sus ainable py ogenic ca bon cap u e and s o age (PyCCS)
based on land- and calo ie-neu al biomass p oduc ion, capi alizing on yield inc eases
induced by biocha -based e iliza ion (BBF) o main ain calo ie p oduc ion while ealizing
ne CO2 emo al om he a mosphe e.
In clima e economic models wi h cos op imiza ion (In eg a ed assessmen models,
IAMs), scena ios compa ible wi h a maximum wa ming o below 1.5 °C o 2 °C equen ly
ely on ex ensi e BECCS deploymen . They p ojec equi ed a es o up o mo e han 9
G CO2 yea −1 a ound he yea 2050 (median: 2.75 G CO2 yea −1), eaching maximum le -
els o mo e han 16 G CO2 yea −1 by 2100 (median: 8.96 G CO2 yea −1, 15 h–85 h pe cen-
ile: 2.63–16.15 G CO2 yea −1) (IPCC 2022). Howe e , he e is la ge skep icism whe he
hese simula ed high deploymen olumes o BECCS can ealis ically be achie ed gi en
economic, poli ical, and echnological cons ain s on he assumed apid scale-up o NETs
(Bedna e al. 2019; Lenzi e al. 2018; Neme e al. 2018). Also, se ious conce ns ha e
been aised ega ding subs an ial en i onmen al and social side e ec s: La ge-scale deploy-
men o BECCS om dedica ed bioene gy c ops would lead o addi ional land deg ada-
ion, compe i ion o land wi h bo h ood p oduc ion and biodi e si y p o ec ion, and could
cause s ong inc eases in human wa e and e iliza ion use, among o he s (Boysen e al.
2017; S enzel e al. 2019). All o hese con ibu e o plane a y des abiliza ion by u he
inc easing he p essu e on plane a y bounda ies cha ac e izing humani y’s sa e ope a ing
space (Heck e al. 2018).
PyCCS is p oposed as an al e na i e biomass-based NET and scalable app oach wi h a
high le el o echnological eadiness and applicabili y ac oss a b oad spec um o usages
Table 1 Lis o abb e ia ions
BBF Biocha -based e iliza ion LCN-PyCCS Land- and calo ie-neu al PyCCS
BC100 Biocha ca bon emaining a e 100 yea s NE Nega i e emissions
BECCS Bioene gy wi h ca bon cap u e and
s o age
NET Nega i e emission echnology
BFT Biomass unc ional ype PBIAS Pe cen bias a e Mo iasi e al. (2007)
DACCS Di ec ai ca bon cap u e and s o age PyCCS Py ogenic ca bon cap u e and s o age
IAM In eg a ed assessmen model YI Biocha -media ed yield inc eases
la/sa Ra io o lea a ea o sapwood a ea
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 3 o 28 34
including di e se ag icul u al sys ems, was e managemen , and ma e ial p oduc ion
(Osman e al. 2022). This NET is based on py olysis, he he mochemical decomposi ion
o biomass a high empe a u es (350–900 °C) in an oxygen-de icien a mosphe e. The
h ee main ca bonaceous py olysis p oduc s can subsequen ly be s o ed in di e en ways o
p oduce NE: as solid biocha in soils o building ma e ial, as bio-oil in deple ed ossil oil
eposi o ies, and as CO2 a e combus ion o pe manen -py ogas in geological s o ages in
e y ad anced echnological se ings (Schmid e al. 2019).
The e m PyCCS has been in oduced o co e he whole ange o seques a ion op ions
a ising om he py olysis p ocess, which howe e di e in hei le el o echnological
eadiness and s o age pe manence (Schmid e al. 2019; We ne e al. 2018). This is essen-
ially di e en o he e minology o BECCS and DACCS (di ec ai ca bon cap u e and
s o age), whe e CCS exclusi ely e e s o p ocessing and s o ing CO2 (IPCC 2018). While
biocha applica ions o soil ha e been p ac iced o cen u ies and esea ched o mo e han
one decade, he combina ion o chemical looping combus ion and py olysis, which would
esul in he mos e icien way o he geological s o age o combus ion p oduc s o pe -
manen -py ogases, has no been es ed widely ye (Schmid e al. 2019). Once deployed,
he geological s o age o p ocessed py ogases can be conside ed pe manen (unless leaked
h ough pe meable aul s o ac u es in he seal) acco ding o he assump ions o he same
p ocesses o BECCS and DACCS.
In case o ca bon seques a ion h ough biocha , howe e , he a e o ca bon di e s
be ween applica ions. High du abili y o biocha ca bon s o age in soils can be a ibu ed o
he de elopmen o used a oma ic s uc u es du ing biomass py olysis (Wang e al. 2016).
These s uc u es ende biocha conside ably less suscep ible o mic obial decomposi ion
in compa ison o esh biomass. To ensu e biocha s exhibi high du abili y, p oduc ion
mus occu a ele a ed empe a u es wi h ex ended esidence imes, p omo ing comple e
ca boniza ion and he o ma ion o used a oma ic s uc u es, indica ed by low H/C a ios
(Ippoli o e al. 2020; Spokas 2010). Es ablished me hodologies quan i ying 100-yea bio-
cha pe sis ence (e.g., IPCC (2019)) mainly ex apola e sho - e m decomposi ion o bio-
cha componen s wi h a lowe deg ee o a oma ici y obse ed unde labo a o y condi ions.
Ye , unce ain ies emain as his alls sho o cap u ing p ocesses explaining millennial
pe sis ence and dynamics in he open en i onmen (Leng e al. 2019). Following hese
quan i ica ion me hods, he ac ion o biocha ca bon emaining in he soil a e 100 yea s
is es ima ed o be a ound 70–80% o H/C a ios below 0.5 and py olysis empe a u es
abo e 450 °C (Camps-A bes ain e al. 2015; IPCC 2019; Lehmann e al. 2021). Ye , he
pe manence o py ogenic ca bon seques a ion would be signi ican ly inc eased when he
biocha is used in building ma e ials.
In his s udy, we solely accoun o biocha seques a ion in soils and i s pa icula co-
bene i in ag icul u e, as applying biocha o a able soils po en ially leads o signi ican
inc eases in ag icul u al yields as well as educed wa e and nu ien demand (Schmid
e al. (2021) and me as udies he ein; Bai e al. (2022)), educing he p essu e on land,
wa e , and e ilize esou ces. Fu he mo e, la ge-scale ubiqui ous biocha seques a ion
in soils migh be a o ed o e indus ial-scale op-down app oaches o NETs because i can
be deployed om small-scale o he la ge-scale (subsis ence o indus ial) and he e o e
migh suppo he UN Sus ainable De elopmen Goals (SDGs). This migh be achie ed
by educing dependencies on ex e nal esou ces, ealizing highe ag oecosys em esilience
and wa e pu i ica ion, as well as deli e ing clean cooking echnology wi h py olyze s ha
can educe biomass demand, as epo ed o biocha in Smi h e al. (2019).
Ye , as holds ue o all biomass-based NETs, he sou ce o he eeds ock is he mos
c i ical ac o o he en i onmen al impac o PyCCS. The land and wa e oo p in s
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 4 o 28
o PyCCS eeds ock p oduc ion a e hus minimal i based on esidues om c opland o
o es y (Wool e al. 2010) bu can be mo e subs an ial i based on dedica ed plan a-
ions (We ne e al. 2018). Un o una ely, he global a ailabili y o c op esidues no
al eady used o o he pu poses is highly unce ain (Hanssen e al. 2020). An in igu-
ing addi ional op ion o sus ainable eeds ock p oduc ion is unique o PyCCS: biomass
inpu om dedica ed as -g owing s ocks p oduced in land-neu ali y. I signi ican le -
els o biocha -media ed yield inc eases (YI) we e achie ed, he same amoun o ood
could be p oduced on less land. Thus, a ac ion o he c opland could be dedica ed
o as -g owing biomass supplying PyCCS eeds ocks wi hou equi ing addi ional land
(Fig.1).
We ne e al. (2022) es ima ed he NE po en ial o LCN-PyCCS (land- and calo ie-
neu al PyCCS) as 0.44–2.62 G CO2 yea −1 depending on he achie able deg ee o YI
abo e p esen le els on (sub-) opical c opland (15–30%) assuming an applica ion a e o
2 ha−1. No e ha he highe end o he ange equi es e y op imis ic assump ions such as
he de elopmen o op imized biocha applica ions adap ed o speci ic soils and c ops (see
below) and/o he inc ease o soil-c op sys em esilience agains ex eme wea he /clima ic
e en s ha s ongly educe ag icul u al p oduc ion.
Howe e , ecen s udies and me a-analyses indica e ha signi ican YI can s ill be
eached wi h lowe applica ion a es (such as < 1 ha−1) i ope a ed as BBF ins ead o
as a gene al soil amendmen . Bulk soil amendmen wi h biocha is he inco po a ion o
pu e, un ea ed biocha o ag icul u al land whe e i is ploughed o d illed in o he soil.
Fig. 1 Schema ic ep esen a ion o land- and calo ie-neu al PyCCS (LCN-PyCCS) indica ing he anges
o he ope a ion space assessed in his s udy (whi e boxes; g een ame: anges o eeds ock managemen ,
blue ame: anges o py olysis p ocess, b own ame: anges o c op yield esponse o biocha -based e i-
liza ion). De ails on he assessmen anges a e gi en in Table S1
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 5 o 28 34
In con as , BBF e e s o ei he biocha - e ilize mix u es (mine al o o ganic) placed
concen a ed in he oo zone (Schmid e al. 2017; Su adha e al. 2021), o g anula /
pelle ized biocha e ilize s ha o en consis o clay/silica e mine als, (mine al) e iliz-
e s con aining ni ogen, phospho us and po assium, and o he nu ien s, plus an un ea ed,
p e- o pos - ea ed ( unc ionalized) biocha componen (Joseph e al. 2021). Wi h BBFs,
compa ably low biocha addi ions o < 1 ha−1 can ha e conside able e ec s (G a mülle
e al. 2022; Qian e al. 2014). The me a-analysis o Melo e al. (2022) epo ed a g and
mean e ec o 10% YI in compa ison o he e ilized con ol a an a e age applica ion a e
o 0.8 ha−1 and e en 17% o cha s wi h a ca bon con en o > 30%.
In compa ison o bulk amendmen wi h biocha , ailo ed BBF could ex end he geo-
g aphic applicabili y o LCN-PyCCS o wo easons: (i) he posi i e yield e ec s o BBF
could also be obse ed in empe a e egions whe eas he amendmen app oach inc eases
yields mos ly only in he (sub-) opics; (ii) he lowe applica ion a es o BBF dec ease he
biocha demand and he eby he yield equi emen s o LCN- eeds ock p oduc ion.
To in es iga e he po en ially ex ended applicabili y o LCN-PyCCS based on BBF, we
quan i y i s global NE po en ial by applying he biogeochemical biosphe e p ocess model
LPJmL o simula e he biomass ha can po en ially be p oduced as py olysis eeds ock
unde his land- and calo ie-neu al app oach. We ex end he analysis u he by add essing
he sensi i i y o LCN-PyCCS po en ials o assump ions abou (i) py olysis p ocess pa am-
e e s, (ii) he managemen in ensi y o he eeds ock p oducing sys em, and (iii) biocha
du abili y in soils. In he case o (i), we conside a ange be ween wo se s o pa ame e s
ep esen ing a conse a i e assump ion and an op imized se ing o accoun o he calcula-
ion’s sensi i i y owa ds assumed p ocess-speci ic biocha yields and ca bon con en s in
he cha . Rega ding (ii), we accoun o wo le els o managemen o eeds ock-p oduc-
ing sys ems o e lec on he po en ial o managemen in ensi ica ion. Fo (iii), we assess a
ange o biocha esidence imes in soils cen e ed a ound a base assump ion o e lec he
unce ain y in ega d o du abili y. Fu he mo e, as he ex en and o e all NE po en ial o
LCN-PyCCS s ongly depends on he biomass yields, he analysis is p eceded by adap -
ing he mos impo an pa ame e s o he ep esen a ion o as -g owing plan s po en ially
used as py olysis eeds ock based on compa isons o simula ed yields and obse a ions.
2 Me hods
LCN-PyCCS is a sys em o land-neu al biomass p oduc ion on c oplands using biocha -
media ed YI o main ain calo ie p oduc ion while ealizing ne CO2 ex ac ion om he
a mosphe e. Th ough he YI, a ac ion o he c opland can be dedica ed o PyCCS eed-
s ock p oduc ion o p o ide sel -su icien biocha supplies and NE while p ese ing le els
o ood p oduc ion (Fig.1). Assuming + 10% YI, o example, would allow 110% calo-
ie p oduc ion on he same a ea o 100% p oduc ion on 91% o he a ea, lea ing 9% o
PyCCS eeds ock p oduc ion. Whe he c opland is sui able o he LCN-PyCCS app oach
he e o e depends on he po en ial biomass p oduc ion on he ededica ed land. Only i he
biomass yield p o ided enough eeds ock o supply he emaining c opland wi h su icien
biocha (i.e., 0.8 ha−1 yea −1 mean in Melo e al. (2022)) o main aining he calo ies
p oduced, a ac ion o he land would be conside ed o biocha eeds ock p oduc ion. Ye ,
his is a conse a i e assump ion, because i does no include (a ac ion o ) he c op esi-
dues, which a e in p ac ice o en added o biocha p oduc ion, e.g., by smallholde a me s
(Schmid e al. 2017).
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 6 o 28
2.1 The global biosphe e model LPJmL
A spa ially explici es ima e o po en ial biomass p oduc ion is equi ed o an assessmen
o global heo e ical po en ials o he LCN-PyCCS app oach. In his s udy, we apply he
p ocess-based global biogeochemical ege a ion model LPJmL ( e sion 4.0) o simula e
he g ow h o dedica ed PyCCS eeds ocks (lignocellulosic g asses and as -g owing ee
species) wi h a daily ime s ep and a spa ial esolu ion o 0.5° × 0.5°. Simula ing key
ecosys em p ocesses in di ec coupling o he ca bon and hyd ological cycle, he model
es ima es he ege a ion dynamics o 11 na u al plan unc ional ypes (Si ch e al. 2003)
and 12 c op unc ional ypes and managed g assland (Bondeau e al. 2007); o de ailed
desc ip ions and alida ions o he biogeochemical dynamics, see Schapho e al. (2018a)
and Schapho e al. (2018b). Addi ionally, h ee ypes o second gene a ion ene gy c ops
a e included (biomass unc ional ypes; BFTs) o es ima e po en ial eeds ock p oduc ion
o biomass-based NETs: wo as -g owing ee species o woody biomass pa ame e ized
as eucalyp in opical clima es and popla and willow in empe a e clima es and lignocel-
lulosic C4 g ass o he baceous ene gy c ops (Be inge e al. 2011; Heck e al. 2016).
2.2 Sensi i i y analysis o pa ame e s o simula ing biomass p oduc ion
We p ima ily calcula e he LCN-PyCCS po en ial based on he baceous eeds ock o ensu e
annual biomass supply. This ocus has been es ablished in p io s udies on global es i-
ma es, i.e., We ne e al. (2018) and We ne e al. (2022), because he g assy BFT shows
highe yields in LPJmL, and biocha s om he baceous eeds ock o en ha e be e yield-
inc easing p ope ies han woody biocha s; see me a-s udies compiled in Schmid e al.
(2021). Sys ems o wood ha es and sho o a ion coppice a e ypically ha es ed in a
mul i-annual cycle (Li e al. 2018), which could cause a biocha de ici in he LCN-PyCCS
app oach (i no supplemen ed by esidues o o he biomass sou ces). Howe e , he imple-
men a ion o LCN-PyCCS can be di e se depending on he a m’s condi ions and needs,
whe e he ededica ion o c opland o woody species (e.g., hedge ows) migh be p e e ed
o ecological easons and he biomass de ici migh be balanced h ough annual p uning o
selec i e logging, which is no ep esen ed in he model. To p o ide a i s es ima e o he
LCN-PyCCS po en ial o hese woody eeds ocks, we addi ionally applied ou calcula ions
o he biomass ha es simula ed o he wo woody BFTs in LPJmL, a e aged o e he
plan a ion li e ime.
To ensu e a obus ep esen a ion o BFTs, we in es iga e he sensi i i y o he simu-
la ed yields o a ia ions o selec ed pa ame e s cha ac e izing plan physiology and man-
agemen , which a e mos ele an o he simula ion o biomass p oduc ion in he model.
The g assy BFT ollows g ow h dynamics o opical C4 g ass in LPJmL, ep esen -
ing as -g owing species like Miscan hus and swi chg ass. While he lignin- ich suppo -
i e issue enabling annual ha es s h ough con inuous g ow h ha is cha ac e is ic o ,
i.e., Miscan hus, is no ep esen ed in he model, i can s ill ep esen a highly p oduc i e
g ass unc ional ype op imized owa ds biomass p oduc ion h ough mul iple ha es s pe
yea . In LPJmL, he g assy BFT is ha es ed whene e abo eg ound biomass eaches a
ce ain h eshold and when senescence is eached. The ha es h eshold con ols he in e -
als be ween ha es s and eg ow h dynamics and he eby signi ican ly impac s simula ed
yield le els. Low alues esul in longe g owing pe iods wi h s agna ing p oduc i i y and
he eby lowe yields. Fu he mo e, he yields depend on he ha es index, i.e., he ac-
ion o biomass emo ed. To bes ma ch epo ed annual ha es sums, we e isi hese
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 7 o 28 34
pa ame e s (ha es h eshold o 400 gC m2 and ha es index o 0.75 selec ed in Heck
e al. (2016)) based on a compa ison o simula ed biomass yields wi h a new ex ensi e
obse a ional da ase on bioene gy c op yields (Li e al. 2018). Fo his, we a y he ha -
es h eshold and ha es index in a li e a u e-based ange (150–450 gC m−2 and 0.7–0.9,
espec i ely; S1) and compa e simula ed yields o obse a ions om 90 si es in Li e al.
(2018). We exclude swi chg ass obse a ions because he biomass p oduc ion o Mis-
can hus is signi ican ly highe (Li e al. 2018; Li e al. 2020), which makes a combined
ep esen a ion o bo h species less ele an (Ai e al. 2020) and decisions o g ow he mo e
p oduc i e c op mo e likely (Zhuang e al. 2013).
In case o he woody BFTs, we assess he esponse o a ying he a io o he cha -
ac e is ic a eas o lea es and sapwood (la:sa). The pa ame e la:sa has been iden i ied by
Zaehle e al. (2005) as one o he pa ame e s o plan g ow h ha in luence he p oduc i i y
o ees mos signi ican ly. Howe e , i has no been adap ed o he woody biomass plan-
a ions ye (while he o he impo an pa ame e s we e adjus ed, see S2). As lowe la:sa
alues inc ease he amoun o ca bon equi ed o lea es and associa ed anspo issue,
he eby educing he lea a ea bu enhancing he ca bon s o age in wood, lowe la:sa alues
can be expec ed o species chosen o hei enhanced biomass p oduc ion. He e, we assess
he sensi i i y o simula ed biomass p oduc ion in LPJmL o a li e a u e-based ange o
la:sa alues ( opical: 2500–5000, empe a e: 2000–5500; see S1) and e alua e he espec-
i e model pe o mance acco ding o obse a ions.
While he managemen o biomass plan a ions can a y widely in p ac ice (i.e., e i-
liza ion, pes con ol, soil p epa a ion, i iga ion, e c.), a ia ion in plan a ion manage-
men o BFTs in LPJmL is ep esen ed by cell-speci ic i iga ion ( ep esen ing manage-
men in ensi y) and o woody BFTs by a BFT-speci ic o a ion leng h, i.e., he yea s
o g ow h be o e coppice. While i iga ion can be used o spa ially a y managemen
in ensi y le els o di e en scena ios, he o a ion leng h is p ede ined o each woody
BFT and has been se o 8 yea s o bo h ypes in he o iginal pa ame e iza ion (Be -
inge e al. 2011). Howe e , he o a ion leng h can be qui e a iable in p ac ice wi h a
median o 3 yea s o sho o a ion coppice sys ems o willow o popla and 6 yea s o
eucalyp plan a ions epo ed in he Li e al. (2018) da abase. In combina ion wi h he
ange o la:sa, we assess he model’s esponse o a ying his pa ame e o a ange o
1–12 and 2–10 o he o a ion leng h o opical and empe a e ees, espec i ely, co -
e ing he 10 h o 90 h pe cen ile o plan a ion age (including all expe imen s, empe a e n
= 1068, opical n = 439) and o a ion leng h ( epo ed as common p ac ice, empe a e
n = 678, opical n = 96).
The global yield da ase o majo lignocellulosic bioene gy c ops epo ed o ield
measu emen s compiled by Li e al. (2018) p o ides an ex ensi e da abase o e alua ion
o simula ed biomass yields and he eby p o ides a sui able e e ence o pa ame e selec-
ion based on he pe o med sensi i i y analyses. We simula e he g ow h and ha es o
i iga ed and ain ed plan s unde clima e condi ions o 1985–2014 and calcula e he mean
yields o e i e o a ions o woody ypes and o e 30 yea s o he he baceous ype. These
LPJmL-compu ed mid- ange yields be ween ain ed (no i iga ion) and in ensi ied ( ull
i iga ion) a e hen compa ed o he mean o he minimum and maximum epo ed yields
o expe imen al es si es loca ed in he espec i e g id cell ( a ying in obse a ions pe i-
ods (1968–2016), mean sampling yea : 1999). The model pe o mance is assessed by he
me ic o pe cen bias (PBIAS), he sum o biases di ided by he sum o obse ed alues
(Mo iasi e al. 2007) excluding ou lie s o he ela i e di e ence be ween obse ed and
simula ed yields, de ined as alues below he 25 h pe cen ile minus 1.5*in e qua ile ange
o abo e he 75 h pe cen ile plus 1.5*in e qua ile ange.
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 8 o 28
2.3 Simula ion se ‑up o LCN‑PyCCS scena ios
Fo he assessmen o he heo e ical NE po en ial o LCN-PyCCS, we apply LPJmL o
simula e he g ow h o he BFTs unde a pa ame e selec ion ha is based on he sensi-
i i y analysis and e alua ion o model pe o mance desc ibed abo e.
The model is d i en by clima e inpu om he gene al ci cula ion clima e model
HadGEM2-ES as con ibu ed o he ISIMIP2b ensemble o he RCP2.6 SSP2 pa hways
(F iele e al. 2017) and co esponding CO2 concen a ions as well as da a on soil ex u e
based on he Ha monized Wo ld Soil Da abase (FAO e al. 2012). P eceding he simula-
ions om 2025 o 2099, an ini ial spin-up o 5000 yea s is pe o med o achie e an equi-
lib ium o soil ca bon and dis ibu ion o na u al ege a ion ollowed by 390 yea s o a
ansien spin-up in oducing he in luence o ag icul u e on he ca bon balance wi h his-
o ic land use change un il 2015 based on HYDE 3.2 (Klein Goldewijk e al. 2017).
The ealloca ion o c opland o biomass p oduc ion o PyCCS is based on he land
use p ojec ions o a RCP2.6 SSP2 scena io ealiza ion o he land alloca ion model
MAgPIE (Die ich e al. 2019), p o ided in he ISIMIP2b ensemble ha is consis en
wi h he HadGEM2-ES clima e inpu (F iele e al. 2017). The ac ion o c opland
dedica ed o biocha eeds ock p oduc ion (9%) is based on he assump ion o 10% YI
achie able h ough BBF, co esponding o he g and mean o yield esponses epo ed
in Melo e al. (2022). In he assessmen , we d aw a ange o 5% and 15% YI a ound his
base assump ion (acco ding o he espec i e con idence in e al in Melo e al. (2022))
o accoun o unce ain ies and dependencies in he yield esponse. In addi ion, we es
o a scena io o biocha applica ion op imized owa ds ca bon seques a ion and yield
esponses wi h 20% YI (Fig.1, S1), which is wi hin he ange o he con idence in e al
o biocha wi h a ca bon con en > 30% (CI 11–24%) in Melo e al. (2022).
2.4 Managemen in ensi ies
To analyze he e ec o managemen o eeds ock p oduc ion and esul ing yields on NE
po en ials, we assess wo managemen in ensi ies on he ededica ed c opland (Fig.1, S1).
Fi s , we assume minimal managemen , e lec ing a case whe e he a me ’s managemen
e o s ocus on he emaining c opland. The eeds ock is hen simula ed as ain ed biomass
yields in LPJmL. In he sensi i i y analysis, i iga ion mee ing he o al wa e demand o he
plan a ion ep esen s he uppe end o he ange o ag icul u al managemen . In line wi h his,
we assess a second scena io assuming mode a e managemen as he mid- ange yield be ween
ain ed (no i iga ion) and in ensi ied ( ull i iga ion).
2.5 Py olysis pa ame e s andseques a ion e iciencies
Fo he py olysis p ocess ans o ming he ha es ed biomass in o biocha , we assume pa am-
e e s o slow py olysis wi h a highes hea ing empe a u e o 500 °C o ensu e ela i ely high
biocha yields a he same ime as high ac ions o ecalci an biocha . As NE po en ials o
simula ed biocha applica ions s ongly depend on he assumed p ocess- and eeds ock-spe-
ci ic biocha yields and ca bon con en s in he cha , we s udy wo se s o pa ame e s ep esen -
ing a conse a i e and an op imized se ing, se ing a ange (Fig.1, S1). The i s se shown in
Table2 is based on Wool e al. (2021) by a e aging o e a la ge numbe o di e en py olysis
echnologies, while he second se ep esen s se ings ha a e op imized owa ds biocha p o-
duc ion o ca bon seques a ion ollowing he biocha yield equa ions o Schmid e al. (2019)
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 15 o 28 34
Fig. 3 Cell ac ions dedica ed o LCN-PyCCS in 2099 (a) and annual global sums o nega i e emissions a e aged o e 2025–2099 (b) based on di e en assump ions o
py olysis pa ame e s and managemen o he eeds ock-p oducing sys ems, assuming 10% biocha -media ed yield inc ease. Combina ions o highe po en ial (highes : mode -
a e s ock managemen plus op imized py olysis pa ame e s, g een) include he a ea o combina ions wi h lowe po en ial (lowes : ma ginal s ock managemen plus conse a-
i e py olysis pa ame e s, pu ple). The segmen s o he ba plo s in b ep esen he po en ial unde he assump ion o 74% biocha ca bon emaining in he soil a e 100 yea s,
while he e o ba s show he ange o he lowe (70%) and highe (80%) du abili y es ed
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 16 o 28
illus a i e mi iga ion pa hway o Shi ing Pa hways which assumes a - eaching ans-
o ma ions in socie y and economy (IPCC 2022; Soe gel e al. 2021). The annual PyCCS
a es based on mode a e managemen co espond o 82–115% (conse a i e o op imized
assump ions o py olysis unde BC100 = 74%) o he BECCS demand in his illus a i e
mi iga ion pa hway wi h a ocus on shi ing ansi ion owa ds Sus ainable De elopmen
Goals, including po e y educ ion and b oade en i onmen al p o ec ion in addi ion o
deep GHG emissions cu s. Consequen ly, he e is e idence om IAM assessmen s ha NE
a es o he magni ude quan i ied in his s udy could suppo he clima e a ge s o he Pa is
Ag eemen . This is he case, howe e , only i he s ingen deca boniza ion measu es in
combina ion wi h he comp ehensi e socioeconomic ans o ma ions (e.g., die changes,
minimized ood was e, and imp o ed dis ibu ion o goods) assumed in hese scena ios a e
success ul, and no o he mains eam collec ion o IAM pa hways beyond he illus a i e
pa hway men ioned.
In addi ion o enhancing he py olysis p ocesses and he p oduc i i y o he eeds ock
supply, we es ed a case o imp o ed BBF applica ion leading o 20% yield inc eases, i.e.,
as obse ed o cha s wi h pa icula ly high ca bon con en in Melo e al. (2022) (17% o
> 30% ca bon con en ) and expec ed o ailo ed biocha s (Joseph e al. 2021). This op imi-
za ion on he applica ion side may inc ease he NE p oduc ion o up o 2.45 G CO2 yea −1
unde he assump ion o op imized py olysis pa ame e s, mode a e eeds ock managemen ,
and BC100 = 74%.
While py olysis pa ame e s and eeds ock p oduc ion impac he sui able a ea o he
LCN-PyCCS app oach (Fig. 3a), he p esumed biocha ca bon du abili y in soils d i es
he depic ion o ca bon losses o e 100 yea s, ul ima ely a ec ing he inal seques a ion
po en ial (Fig.3b). A biocha ca bon e en ion a e as low as 70% o he ini ial biocha
ca bon inpu would esul in seques a ion po en ials o 0.19 G CO2 yea −1 o ma ginal
and 0.70 G CO2 yea −1 o mode a e eeds ock managemen in ensi y, conside ing 10% YI
and conse a i e py olysis pa ame e s (Table 3). Unde op imized py olysis condi ions,
wi h his lowe BC100 o 70%, po en ial seques a ion could each 0.35 G CO2 yea −1 o
ma ginal and 1.04 G CO2 yea −1 o mode a e eeds ock managemen in ensi y. Assum-
ing a highe ac ion o biocha ca bon emaining in soil wi h BC100 = 80% would sig-
ni ican ly inc ease he o e all seques a ion po en ial o 0.22 G CO2 yea −1 o ma ginal
and 0.80 G CO2 yea −1 o mode a e eeds ock managemen in ensi y, gi en 10% YI and
conse a i e py olysis pa ame e s. Op imizing he py olysis p ocess in his case could aise
he po en ials up o 0.40 G CO2 yea −1 o ma ginal and 1.19 G CO2 yea −1 o mode a e
eeds ock managemen in ensi y.
As eeds ock ypes o py olysis can be e y di e se and woody inpu s a e also widely
conside ed o biocha p oduc ion (Ye e al. 2020), we addi ionally es ed woody eeds ock
o he LCN-PyCCS app oach conside ing he as -g owing ee unc ional ypes ep e-
sen ed in LPJmL and pa ame e ized in his s udy. We ind ha he NE po en ials a e signi i-
can ly lowe han quan i ied o he g assy eeds ock because he equi ed biomass yields
o su icien biocha supply a e ba ely eached wi h he woody unc ional ypes in he
model. Only i hese biomass p oduc ion sys ems a e mode a ely managed, woody eed-
s ock as ep esen ed in his s udy may become sui able o LCN-PyCCS in a ew highly
p oduc i e egions, esul ing in ela i ely low NE po en ials o 0.10 G CO2 yea −1 and 0.15
G CO2 yea −1 o conse a i e and op imized py olysis pa ame e s, espec i ely.
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 17 o 28 34
4 Discussion
Clima e s abiliza ion is c ucial o Ea h sys em s abili y, bu when biomass-based NETs
ha compe e o land wi h ood p oduc ion and ecosys em p o ec ion a e implemen ed
wi hou conside a ion o he en i onmen al (e.g. land equi ed) and socie al (e.g. calo-
ies p oduced) epe cussions, s abiliza ion measu es may a he same ime h ea en ea h
sys em s abili y by unde mining biosphe e in eg i y. Ins ead o expanding NET deploy-
men in o de o each a ce ain NE a ge as in op imiza ion models o clima e economics,
his assessmen quan i ied he NE supply achie able “bo om-up” unde he cons ain s o
land- and calo ie-neu ali y. While s aying wi hin he bounds o c opland and main ain-
ing calo ie p oduc ion, LCN-PyCCS based on BBF may p oduce 0 o 2.03 G CO2 yea −1
(0.19–1.19 G CO2 yea −1 o he base assump ion o + 10% YI; 0–0.01 G CO2 yea −1 o
+ 5%; 0.57–2.03 G CO2 yea −1 o +15%) depending on (i) he YI achie ed h ough BBF,
(ii) he py olysis pa ame e s assumed, (iii) he managemen in ensi y o he biomass p o-
ducing sys em, and (i ) he biocha du abili y in soils. We a gue ha in o de o es ima e
ealis ic po en ials o NE, he disc epancy ound be ween he demand-d i en NE olumes
calcula ed in economic op imiza ion models (IAMs) o pa hs eaching ambi ious clima e
a ge s and he esul s o supply-d i en app oaches as assessed he e needs o be anspa -
en ly discussed.
Ou esul s suppo he indings in We ne e al. (2022) ha global LCN-PyCCS po en-
ials a e d i en by he biocha -media ed YI ha can be accomplished. A he lowe ange,
we ind ha a le el o + 5% YI is no su icien o he LCN-PyCCS app oach based on
BBF. Besides he high eeds ock yield equi emen s o la ge biomass subs i u ion demand,
such low yield esponses a e no likely o encou age ededica ion o land o eeds ock
p oduc ion. In cases whe e biocha applica ion is s ill p e e ed despi e low YI (i.e., o
enhanced soil esilience), eeds ock would hus need o be supplied by o he (sus ainable)
sou ces.
In ou assessmen , we employed he insigh s om Melo e al. (2022) o subs an ia e
uni e sal YI le els in he adop ion o a sys ema ic me hodology. Howe e , he ac ual
yield imp o emen achie able ac oss di e se loca ions migh su pass o all sho o his
alue, because he esponse o BBF exhibi s conside able a iabili y, as e idenced by he
da a compiled by Melo e al. (2022). A mo e p ecise, spa ially explici compu a ion o YI
unde cu en condi ions would necessi a e he in eg a ion o di e se ac o s, among o he s
encompassing soil ca ego y, e ilize ype, and c op a ie y. Ye , he cu en ly a ailable
da a is no su icien o a s a is ical model o ha kind. In he Melo e al. (2022) da a-
se , he obse a ions associa ed wi h one ca ego y o an explana o y a iable can show a
subs an ial ange in BBF esponse; o ins ance, he ca ego y o “weakly de eloped soils”
exhibi s a con idence in e al spanning om 1 o 25% YI. Addi ionally, esponses ac oss
ca ego ies wi hin a a iable, like “weakly de eloped soils” and “highly wea he ed soils,”
migh no exhibi signi ican di e ences. While a comp ehensi e s a is ical model o
de i ing BBF-induced yield inc ease h ough mul iple explana o y a iables is p esen ly
absen , ou s udy ollows a sys ema ic app oach wi h heo e ically uni e sal le els o YI
o e lec on magni udes o he seques a ion po en ial o LCN-PyCCS and i s sensi i i y
o achie able YI, managemen in ensi y, py olysis pa ame e s, and s o age du abili y. Ye ,
as a eas iden i ied as sui able o LCN-PyCCS a e no dis ibu ed equally, we ha e e i ied
ha he designa ed egions o LCN-PyCCS p edominan ly coincide wi h soil o de s ha
ha e shown signi ican esponse o BBF (S4). Beyond he e alua ed dis ibu ion o LCN-
PyCCS, o he egions o ag icul u al p oduc ion a e la gely ma ked by mo e p onounced
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 18 o 28
esponses o e iliza ion and/o soils di e en om highly wea he ed and weakly de el-
oped soils. I needs o be no ed ha hese dis inc soil cha ac e is ics exhibi ed less signi i-
can o no yield esponses o biocha addi ion in he me a-analysis by Melo e al. (2022).
Thus, i BBF applica ions we e assumed o concen a e on hese less esponsi e soils, he
o e all ne impac could be signi ican ly educed.
In addi ion o a ocus on esponsi e soils, we assume eeds ock ea men du ing py oly-
sis ha maximizes ca bon s o age ia biocha . Consequen ly, he en isioned BBF in ol es
biocha cha ac e ized by ele a ed ca bon con en (conse a i e es ima e: 63% and op i-
mized es ima e: 67%). This aligns wi h he subse o cha s in he > 30% ca bon con en
ca ego y, which demons a ed no ably ele a ed yield esponses in he Melo e al. da ase
(mean YI: 17%).
Enhancing he biocha applica ion in e ms o s onge esponses o BBF in plan p o-
duc i i y (also by combining i wi h ano he land-based NET such as enhanced wea he -
ing) may signi ican ly inc ease he NE po en ial. Resea ch and de elopmen on BBFs a e
picking up pace, alongside wi h a g owing unde s anding o biocha -su ace in e ac ions
wi h majo nu ien s such as ni ogen and p e- and pos -p oduc ion ea men op ions o
inc ease desi ed e ec s o c op yields, inc eased ni ogen use e iciency, and educed
en i onmen al ni ogen pollu ion (s a e o knowledge e iewed in Rasse e al. (2022)). An
example is he in il a ion o he po ous biocha s uc u e wi h mol en u ea, p o iding a
slow- elease compound biocha e ilize (Wang e al. 2021; Xiang e al. 2020), he coa -
ing o con en ional e ilize s wi h biocha o inc ease he ni ogen use e iciency (Jia e al.
2021), biocha su ace oxygena ion o inc ease ammonia (NH3) so p ion o acid ea men s,
and o ganic coa ing o inc ease ni a e cap u e (Rasse e al. 2022) and o he s a egies. I
can hence be expec ed ha he uppe ceiling o YI achie ed wi h ailo ed BBFs has no ye
been eached.
Beyond he impac o accomplishable YI, ou esul s indica e a s ong sensi i i y o he
NE po en ials o he assumed py olysis pa ame e s. Basing he calcula ions on he op i-
mized pa ame e se ins ead o he conse a i e assump ion inc eased he NE po en ial
by 40–75%. Thus, p ac i ione s aiming o ca bon seques a ion should ollow se ings o
hei py olysis plan /kiln ha enhance he biocha yield and conside ash o ock powde
supplemen s (Buss e al. 2022; Mašek e al. 2019); ock powde (enhanced wea he ing)
ep esen s ano he NET ha can also inc ease yields (Bee ling e al. 2020; Kan zas e al.
2022).
Fu he mo e, we iden i ied he managemen o he biomass supplying sys ems as
ano he ac o d i ing he NE po en ials in his assessmen . The NE p oduc ion could be
inc eased by + 200–270% i he managemen o biomass p oduc ion was assumed o be
in ensi ied om ma ginal o mode a e le els. Ele a ed yields do no only inc ease he NE
p oduc ion pe hec a e, bu also expand he a ea ha is sui able o he app oach because
mo e egions mee he biomass p oduc ion on ededica ed land equi ed o supply su i-
cien biocha o he c opland. This is pa icula ly ele an o he expansion in he sub-
opics o e en empe a e egions ha only become sui able o he LCN-PyCCS app oach
in es iga ed in his assessmen when mode a e managemen is assumed. Ye , in mos o
he sub opical and empe a e egions, such a deg ee o in ensi ica ion is also mo e likely
because he ag onomic de elopmen in hese coun ies al eady p o ides he in as uc u e
and esou ces equi ed.
A he same ime, he empe a e egions a e also bes ep esen ed in he obse a ional
da ase ha was used o he compa ison wi h simula ed yields, d i ing he pa ame-
e selec ion in his analysis. While he PBIAS could be inc eased signi ican ly wi h he
new pa ame e iza ion o all BFTs, we also iden i ied sho comings in ep esen ing plan
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 19 o 28 34
physiology. Fu he mo e, he model pe o mance could only be assessed whe e obse a-
ion da a was p o ided. Ye , o he opics, he e is a signi ican lack o da a on lignocel-
lulosic ene gy c ops in he li e a u e (Fig.2, Li e al. 2018). We he e o e conclude ha
in o de o ex en he obus ness o he ep esen a ion o such c ops in global modeling,
bo h, he simula ed p ocesses o plan physiology and he geog aphic co e age o alida-
ion da a, would need o be enhanced.
As we es ed he quan i ica ion o LCN-PyCCS o woody biomass by applying as -
g owing ee unc ional ypes as ep esen ed in LPJmL o he eeds ock supply, we cal-
cula ed NE po en ials ha we e signi ican ly lowe han o he assessmen o g assy eed-
s ock. Howe e , hese unc ional ypes ail o ep esen all possible wood-like eeds ocks
ha could be used o LCN-PyCCS (e.g., co ee, ea, cocoa, o ui /nu ee p uning
wood). Mo eo e , depending on he clima ic, soil, and managemen condi ions o he c op-
land, as -g owing ee species (e.g. ag o o es y sys ems) can be mo e bene icial o plan
g ow h, soil esilience, and in e media e- e m landscape ca bon seques a ion (Dollinge
and Jose 2018; Lo enz and Lal 2018) and migh hus be p e e ed o e g assy species, o
be combined wi h hem. Resul ing po en ial bene i s and si e-speci ic e ec s like nu ien
cycling, shade co e , inc easing ela i e ai humidi y, and oo g ow h p e en ing soil e o-
sion (Fahad e al. 2022; To alba e al. 2016) a e no ep esen ed in ou model and hus did
no con ibu e o his assessmen , bu would ein o ce a he han comp omise ou indings.
In addi ion o eeds ock p oduc ion and he pi o al ole o py olysis condi ions in de e -
mining he inpu o biocha ca bon in o he soil, we also assessed a ange o al e na i e
assump ions o he biocha du abili y in soils o accoun o he unce ain y associa ed
wi h his aspec . I is known ha biocha du abili y is enhanced by a g ea e p opo ion
o used a oma ic s uc u es, which o m a highe empe a u es and wi h ex ended esi-
dence imes (Ippoli o e al. 2020; Wang e al. 2016). Howe e , es ima ing he po ion ha
emains a e , o ins ance, 100 yea s is elian on es ima ions, gi en he absence o long-
e m expe imen s (Leng e al. 2019). Cu en me hodologies o quan i ying 100-yea bio-
cha pe sis ence, such as hose used in IPCC in en o y guidelines, p ima ily ex apola e
sho - e m soil decomposi ion p ocesses. These me hods ail o en i ely encompass he
mechanisms con ibu ing o millennial pe sis ence. Consequen ly, he ac ual biocha ca -
bon esidence ime migh exceed hese es ima ions. Concu en ly, u u e esea ch should
ocus on biocha incuba ion in ield condi ions, aiming o highligh dis inc ions om labo-
a o y se ings (Leng e al. 2019). This ield esea ch may unco e ac o s ha could in en-
si y decomposi ion compa ed o labo a o y expe imen s.
Fu he mo e, he s udy is limi ed o py ogenic ca bon seques a ion and does no con-
side o he biocha -media ed p ocesses in he soils ha could po en ially con ibu e o
shi ing he land use sec o om a g eenhouse gas sou ce in o a sink. Biocha -en iched
soils ha e been shown o enhance he build-up o soil o ganic ca bon (Bai e al. 2019;
Blanco-Canqui e al. 2020; Weng e al. 2017) and educe soil acidi y (Singh e al. 2017),
ni a e leaching (Bo cha d e al. 2019; Hagemann e al. 2017), and N2O and CH4 emissions
(Bo cha d e al. 2019; He e al. 2017; Je e y e al. 2016). In o al, biocha ea men s can
hus enhance soil quali y, cu down managemen cos s, and lowe ag icul u al g eenhouse
gas emissions (Kammann e al. 2017; Lehmann e al. 2021).
While we assessed he po en ial o pu pose-g own py olysis eeds ock in he LCN-
PyCCS app oach, i needs o be emphasized ha ha es esidues and was e can p o ide addi-
ional sus ainable biomass inpu s, pa icula ly because o he a oided land compe i ion wi h
ood and na u e. These addi ional sou ces o biocha p oduc ion may e en expand he a ea
o LCN-PyCCS applicabili y as hey could supplemen he biocha supply in egions whe e
he p oduc ion by pu pose-g own eeds ock on ededica ed land alls below he h eshold o
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 20 o 28
su icien biocha applica ion. Ye , he a ailabili y o esidues and was e is unce ain and lim-
i ed (Hanssen e al. 2020; IPCC 2022), and hey canno be conside ed eely a ailable in he
ligh o compe ing uses o bene i s, such as soil ca bon buil -up agains land deg ada ion o
he eplacemen o animal eed (Kal e al. 2020; an Zan en e al. 2018). Ano he sou ce
ha should ecei e mo e a en ion as he biomass ma ke comes unde inc easing p essu e
is eeds ocks ha a e o en o e looked, o example, annual p uning wood o unk and oo
weeding wood (a eplan ing cycles) o sub opical o opical annual and pe ennial c ops
such as co ee, ea, cocoa, palm oil ees, banana, o ui /nu ees. Fo example, was e s eam
and hea demand o he wo k low a co ee p oduc ion si es would a o py olysis (i) o d y-
ing co ee beans in a con olled manne o inc ease p oduc quali y and educe he dange o
ungal con amina ion and (ii) o compos co ee che y pulp wi h biocha . Using such a p od-
uc as a soil amendmen a eplan ing has been shown o accele a e young ee g ow h and
sho en he ime un il he i s ha es is gained (Neumann Co ee G oup, H. Faessle , p ac i-
cal ials, pe s. comm.); biocha -compos s can imp o e soil e ili y o demanding c ops such
as co ee in pa icula in opical soils (Zhao e al. 2020).
This global assessmen aims o illus a e po en ial magni udes o NE achie able h ough
LCN-PyCCS, along wi h i s sensi i i y o speci ic in luencing ac o s. We acknowledge
howe e ha b oad-scale gene aliza ions canno ully encapsula e he unique local condi-
ions ha ul ima ely d i e he seques a ion p ocess. Es ablished me hodologies employed o
e alua e he seques a ion po en ial wi hin speci ic sys ems a e li ecycle assessmen s (LCAs)
in eg a ing all in o ma ion abou inpu s and ou pu s along he pa icula p oduc ion and s o -
age chain (Tisse an and Che ubini 2019). The NE ou come is con ingen upon a mul i ude
o a iables ac oss he li ecycle s ages, including eeds ock selec ion, py olysis condi ions,
anspo a ion dis ances, and ene gy/ uel mix (Azzi e al. 2021). While LCAs se e as alu-
able ools o assessing he CDR po en ial o pa icula sys ems and hei economic iabil-
i y, o en, he speci ic assump ions hey en ail a e no ex apola able o la ge-scale global
scena ios. Consequen ly, hey may ocus on e ining ca bon seques a ion wi hin biocha
p oduc ion unde speci ic ci cums ances. Fo ins ance, Fawzy e al. (2022) demons a ed
his by op imizing he py olysis o oli e ee p uning esidue, yielding a o al seques a ion
o 2.69 CO2-equi alen pe on o biocha . Compa ed o ou s udy, he NE po en ial pe uni
biocha is hus app oxima ely 1.7- o 1.9- old highe and can be a ibu ed o hei op imized
py olysis p ocess, ailo ed o maximum ca bon seques a ion om a lignin- ich eeds ock.
No ably, he composi ion o biocha in he Fawzy e al. s udy, cha ac e ized by a ca bon
con en o 84.9%, BC100 o 92%, and CO2-equi alen expendi u es along he li ecycle o 7%,
s ands in con as o ou assessmen , whe e hese igu es a e 62.7%, 74%, and 18%, espec-
i ely. Ye , o es ima e he magni ude o global po en ials, mo e gene alized assump ions—
albei no aligned wi h he ene s o LCA—become necessa y. Consequen ly, we assessed
py olysis pa ame e anges ollowing unc ions based on obse a ions o he b oade ca -
ego y o g assy (and woody) eeds ock (G a mülle e al. 2022; Schmid e al. 2019; Wool
e al. 2021). Mo e gene ic assump ions we e also made o he ca bon expendi u e h ough-
ou he li ecycle which co esponds o a mode a ely deca bonized ene gy and uel blend,
conside ing an a e age global anspo dis ance o 55 km (see S3).
As desc ibed abo e, ou global assessmen e eals se e al limi a ions ac oss a ious
dimensions: In e ms o eeds ock supply, he a ailabili y o obse a ional da a o ligno-
cellulosic g ass g ow h in opical egions emains spa se. Addi ionally, LPJmL cu en ly
lacks ep esen a ion o mos sui able wood-like eeds ocks, especially hose de i ed om
in eg a ed sys ems like ag o o es y. Fu he mo e, his assessmen does no inco po a e
o he land-neu al esou ces, such as esidues and was e, as biomass inpu s. Mo eo e ,
he global assump ions conce ning yield inc ease and ca bon expendi u e ely on global
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 21 o 28 34
a e ages a he han accoun ing o he speci ic e iciencies o indi idual sys ems. Rega d-
ing he NE po en ial, he assessmen does no inco po a e o he bene i s linked o ca bon
seques a ion ha may a ise om he applica ion o biocha o soils. Ye , despi e hese con-
s ain s, sys ema ic app oaches like his s udy can con ibu e o ad ancing he discou se on
nega i e emissions s a egies, as hey o e a means o assess he magni ude and espon-
si eness o global LCN-PyCCS po en ials conce ning di e se pa ame e s.
Building on he i s global quan i ica ion o LCN-PyCCS po en ials based on biocha
soil amendmen s o 2 ha−1, his s udy does no only add esul s o he p omising p ac ice
o BBF a lowe applica ion a es (a ound 0.8 ha−1), bu also insigh s on he sensi i i y
o he esul s owa ds assump ions on py olysis pa ame e s and he managemen in ensi y
o eeds ock-p oducing sys ems. While he annual a e o 0.44 G CO2 yea −1 epo ed o
+ 15% YI a 2 ha−1 biocha applica ion in We ne e al. (2022) was solely based on op i-
mized py olysis pa ame e s and ma ginal managemen , his analysis illus a es an ope a ion
space o di e en py olysis p ocesses and managemen in ensi ies. Ou calcula ion o he
BBF-based NE po en ial ollowing op imized py olysis pa ame e s and ma ginal manage-
men esul s in 0.37 G CO2 yea −1 as we assume a YI o + 10% acco ding o he g and
mean epo ed in he BBF me a-analysis by Melo e al. (2022). Assuming + 15% YI in
he BBF se ing howe e shows highe NE po en ial (0.82 G CO2 yea −1, Table3) han in
We ne e al. (2022). This can be explained by he lowe applica ion a e ha leads o a
lowe yield h eshold o biomass p oduc ion on he ededica ed land equi ed o su icien
biocha supply on he emaining c opland. Consequen ly, he applica ion can be expanded
in o less p oduc i e egions ha only hen become sui able o LCN-PyCCS. Fu he mo e,
he assump ions on he py olysis p ocess we e e ised, now e e ing o 500 °C ins ead o
450 °C o he highes hea ing empe a u e. This is close o p ac ice and leads o a lowe
biocha yield bu longe esidence imes in he soil, which we conside a mo e easonable
balance o long- e m ca bon seques a ion. To inc ease he obus ness o he assump ions
abou py ogenic ca bon esidence ime in he soil, we addi ionally adjus ed he ac ion o
ca bon ha emains a e 100 yea s om 90% (We ne e al. 2022) o 74% based on ind-
ings by Camps-A bes ain e al. (2015). In ligh o he la es analyses and new e idence on
he ecalci an na u e o biocha p oduced a his empe a u e, he assumed ac ion can be
conside ed conse a i e (Azzi e al. 2024; Sanei e al. 2024).
Ye , add essing he di e ences be ween biocha used as soil amendmen o BBF and
he sensi i i y owa ds py olysis pa ame e s und managemen in ensi ies o he biomass
supply, we conclude ha biogeochemical e alua ions o he global LCN-PyCCS po en ial
can only p o ide a ange o es ima es ha ma k he ou e bounds o he maximum po en ial
unde he gi en assump ions. I is highly unlikely ha a uni e sally uni o m le el o YI
will be achie ed a a global scale. Such a concep is employed he ein solely as a heo e i-
cal assump ion, acili a ing an explo a ion o seques a ion po en ials and pa ame e sensi-
i i ies. The usage and design o he py olysis plan /kiln as well as he biocha applica ion
me hods and yield esponses will always be speci ic o he explici needs and condi ions a
he espec i e p oduce . Ou indings o signi ican ly highe NE po en ials wi h op imized
py olysis pa ame e s and biocha applica ion show ha applying he cons an ly expand-
ing knowledge on op imal ca bon seques a ion and yield esponse in p ac ice can inc ease
he global po en ial o a la ge deg ee. These scena ios a e la gely heo e ical by assum-
ing global applica ions; in ha sense, hey a e pu ely analy ical. Achie able eal-wo ld
po en ials depend on he easibili y o implemen a ion (e.g., cul u al, social, o poli ical
up ake o esis ance), su icien incen i es o in eg a ing PyCCS in o ag icul u al p ac ices
wo ldwide, and es ablished p oo o success ul ope a ion o NE ce i ica ion. Add essing
po en ial esis ance can be acili a ed h ough adhe ence o obus accoun ing egula ions,
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 22 o 28
in eg a ed wi hin comp ehensi e measu emen , epo ing, and e i ica ion amewo ks,
which PyCCS al eady exhibi s p omising en y poin s o (Lehmann e al. 2021). A com-
pelling illus a ion o s ingen quali y con ol o he chemical composi ion o biocha is,
o example, al eady p ac iced wi h he Eu opean Biocha Ce i ica e (EBC 2023; Fawzy
e al. 2021). Fu he mo e, o ci cum en compe i ion o land and esou ces, he implemen-
a ion o PyCCS should concen a e on accoun ing o he wide spec um o co-bene i s,
pa icula ly wi hin he ag icul u al sec o (see abo e), while concu en ly in eg a ing i
wi hin c oss-sec o al planning a he han conside ing i an isola ed clima e change mi i-
ga ion measu e. Fu he mo e, one should no conside indings as ound he e o be ye a
su icien basis o planning on mi iga ion pa hs ha assume hese po en ials will be a ail-
able. Howe e , i is a pa icula s eng h o LCN-PyCCS ha his NET can (and al eady
is) adop ed by di e se ag icul u al sys ems and con ibu es o NE in a bo om-up dynamic,
a he han op-down app oaches like BECCS ha likely ely on cen alized ma ke s and
massi e in es men s o cos ly in as uc u e.
While i is he e o e challenging o make global assump ions on he NE p oduc ion
po en ial o LCN-PyCCS sys ems, we s ill conside i illumina ing o place he global es i-
ma es quan i ied in ou analyses wi hin he ongoing discussion o he la ge-scale imple-
men a ion o NETs o clima e s abiliza ion—e en wi h he la ge anges we ind. Only
h ough such compa isons, a he same scale, can he disc epancies be iden i ied be ween
he la ge NE olumes calcula ed in demand-d i en op imiza ion models and he po en ials
om analyses ocusing on supply-d i en app oaches, cha ac e ized by p econdi ions like
minimizing he p essu e on land esou ces and calo ie p oduc ion.
The a gumen o compa abili y holds ue o he ime scale o echnological de el-
opmen : while PyCCS and pa icula ly biocha assessmen s a e usually based on pa am-
e e s and p ocess unde s anding ha is di ec ly de i ed om ope a ing plan s and eal-
wo ld applica ions, he e alua ions o BECCS and DACCS ypically ollow assump ions
o u u e de elopmen based on only sca ce e idence a comme cial scale (Cha e jee
and Huang 2020; Haikola e al. 2019). The O ca acili y in Iceland is he i s DACCS
plan o ope a e a a comme cial scale wi h a capaci y o 4000 CO2 yea −1 (Ca b ix
2021). In con as o his, he Eu opean Biocha Indus y Conso ium (EBI 2022) epo s
abou 53,000 biocha p oduc ion capaci y buil in 2022 (i.e., PyCCS) in Cen al Eu ope
alone (EBI 2023), po en ially seques e ing abou 90,000–110,000 CO2 based on he
seques a ion e iciencies used o soil applica ions in his s udy. Plan cons uc ion
was p ojec ed o inc ease he Eu opean biocha p oduc ion o abo e 90,000 in 2023
(EBI 2023). Fo BECCS, he de elopmen is ocused on No h Ame ica ha coun s ou
BECCS plan s cap u ing ≥ 100,000 CO2 plus he Deca u plan wi h po en ially one
million CO2 cap u e pe yea . In Eu ope, he DRAX acili y in he UK, wi h a capac-
i y o 330 CO2 cap u e pe yea , is cu en ly he only BECCS plan in ope a ion, ye
se e al p ojec s a e planned (Faja dy 2022; Shahbaz e al. 2021). Thus, he eal-wo ld
ca bon seques a ion o PyCCS in Eu ope exceeds ha o DACCS and BECCS a he
momen . Also, on a global scale, PyCCS (pa icula ly biocha seques a ion) is showing
he highes numbe o ope a ing plan s and ca ches up in ega d o capaci y, when com-
pa ed o BECCS and DACCS (no e ha a global quan i ica ion o biocha p oduc ion is
s ill lacking and would esul in much highe numbe s).
Ou analysis in eg a ing biocha -media ed yield inc eases in he scena io de elopmen
o global NET deploymen iden i ies a po en ial o land- and calo ie-neu al NE p oduc-
ion. The LCN-PyCCS app oach could hus con ibu e o a b oade NET assessmen con-
side ing c i ical addi ional limi a ions in en i onmen al, social, and policy dimensions.
Fu he mo e, as biocha applica ion o soils po en ially enhances soil p ope ies in di e se
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
Page 23 o 28 34
ways (see abo e), he ole o PyCCS could be enhanced in such impac -sensi i e analyses
i mo e biocha -media ed p ocesses (i.e., liming, wa e -holding capaci y, enhanced mic o-
bial ac i i y, inc eased nu ien e en ion) we e ep esen ed in he models. Ye , in pa allel
o hose e o s in p ocess ep esen a ion wi hin ege a ion models, mo e elabo a e models
and da abases o esidue and was e a ailabili y (o bene icial was e-s eam managemen
s a egies in ol ing biocha ) should be de eloped, as es ima es and p ojec ions o sus ain-
able biomass sou ces a e c ucial. While he LCN-PyCCS app oach assessed he e may con-
ibu e o some deg ee o global (mo e sus ainable) NE p oduc ion, a much highe po en ial
could be unlocked i he ou s anding ea u e o PyCCS ha a a ie y o ma e ials can be
p ocessed we e o be exploi ed a a la ge scale, pa icula ly using esidues and was es ha
a e o no o he use (o whe e biocha p oduc ion and use in o ganic was e managemen
o e u he SDG-suppo ing bene i s).
In explo ing a ious al e na i es, i is c ucial o p e en biomass-based NETs—
exclusi ely in ended o clima e s abiliza ion— om becoming a majo d i e o de -
imen al u u e land use change. This conce n is pa icula ly pe inen conside ing he
al eady subs an ial deg ada ion o he Ea h’s biosphe e. In his s udy, we speci ically
highligh BBF-based LCN-PyCCS as one po en ial a enue bu unde sco e he sensi i -
i y o po en ial global NE p oduc ion o ac o s such as achie able YI, py olysis se ings,
and he managemen in ensi y o eeds ock p oduc ion. While ou assessmen add esses
he need o s ic p econdi ions o la ge-scale NET deploymen (i.e., land and calo ie
neu ali y), he indings e eal ha he esul ing po en ials a e subjec o la ge emaining
unce ain ies. Gi en hese unce ain ies, he s udy ein o ces he impe a i e o p io -
i izing deep emission educ ions as he o emos s a egy in clima e s abiliza ion e o s.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps:// doi.
o g/ 10. 1007/ s11027- 024- 10130-8.
Funding Open Access unding enabled and o ganized by P ojek DEAL. This p ojec has ecei ed und-
ing om he Eu opean Union’s Ho izon 2020 esea ch and inno a ion p og am unde g an ag eemen
No 869192 (C.W, J.B.). BMBF BioCAP-CCS p ojec (G an No. #01LS1620A and B). C. K. ecei ed
FACCE-JPI unding o de eloping biocha -based e iliza ion app oaches (p ojec ABC4Soil, G an No.
031B0588B).
Da a a ailabili y Da a suppo ing he main indings o his s udy a e a ailable ia 10.5281/zenodo.7116841.
Land use and clima e inpu da a can be downloaded om he ISIMIP eposi o y, h ps:// da a. isimip. o g/
(F iele e al. 2017) and 10.48364/ISIMIP.208515 (Lange and Büchne 2017).
Decla a ions
Compe ing In e es The au ho s decla e ha hey ha e no con lic o in e es .
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Com-
mons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle
a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he
ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly
om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Mi ig Adap S a eg Glob Change (2024) 29:34
1 3
34 Page 24 o 28
Re e ences
Ai Z, Hanasaki N, Heck V, Hasegawa T, Fujimo i S (2020) Simula ing second-gene a ion he baceous bio-
ene gy c op yield using he global hyd ological model H08 ( .bio1). Geosci Model De 13(12):6077–
6092. h ps:// doi. o g/ 10. 5194/ gmd- 13- 6077- 2020
Azzi ES, Ka l un E, Sundbe g C (2021) Assessing he di e se en i onmen al e ec s o biocha sys ems: an
e alua ion amewo k. J En i on Manag 286:112154. h ps:// doi. o g/ 10. 1016/j. jen m an. 2021. 112154
Azzi ES, Li H, Cede lund H, Ka l un E, Sundbe g C (2024) Modelling biocha long- e m ca bon s o age
in soil wi h ha monized analysis o decomposi ion da a. Geode ma 441:116761. h ps:// doi. o g/ 10.
1016/j. geode ma. 2023. 116761
Bai SH, Omid a N, Galla M, Kämpe W, Tahmasbian I, Fa a MB, Singh K, Zhou G, Muqadass B, Xu
C-Y, Koech R, Li Y, Nguyen TTN, an Zwie en L (2022) Combined e ec s o biocha and e ilize
applica ions on yield: a e iew and me a-analysis. Sci To al En i on 808:152073. h ps:// doi. o g/ 10.
1016/j. sci o en . 2021. 152073
Bai X, Huang Y, Ren W, Coyne M, Jacin he P-A, Tao B, Hui D, Yang J, Ma ocha C (2019) Responses
o soil ca bon seques a ion o clima e-sma ag icul u e p ac ices: a me a-analysis. Glob Chang Biol
25(8):2591–2606. h ps:// doi. o g/ 10. 1111/ gcb. 14658
Bedna J, Obe s eine M, Wagne F (2019) On he inancial iabili y o nega i e emissions. Na Commun
10(1):1783. h ps:// doi. o g/ 10. 1038/ s41467- 019- 09782-x
Bee ling DJ, Kan zas EP, Lomas MR, Wade P, Eu asio RM, Ren o h P, Sa ka B, And ews MG, James
RH, Pea ce CR, Me cu e J-F, Polli H, Holden PB, Edwa ds NR, Khanna M, Koh L, Quegan S, Pidg-
eon NF, Janssens IA e al (2020) Po en ial o la ge-scale CO2 emo al ia enhanced ock wea he ing
wi h c oplands. Na u e 583(7815):242–248. h ps:// doi. o g/ 10. 1038/ s41586- 020- 2448-9
Be inge T, Luch W, Schapho S (2011) Bioene gy p oduc ion po en ial o global biomass plan a ions
unde en i onmen al and ag icul u al cons ain s. GCB Bioene gy 3(4):299–312. h ps:// doi. o g/ 10.
1111/j. 1757- 1707. 2010. 01088.x
Blanco-Canqui H, Lai d DA, Hea on EA, Ra hke S, Acha ya BS (2020) Soil ca bon inc eased by wice he
amoun o biocha ca bon applied a e 6 yea s: ield e idence o nega i e p iming. GCB Bioene gy
12(4):240–251. h ps:// doi. o g/ 10. 1111/ gcbb. 12665
Bondeau A, Smi h PC, Zaehle S, Schapho S, Luch W, C ame W, Ge en D, Lo ze-Campen H, Mulle C,
Reichs ein M, Smi h B (2007) Modelling he ole o ag icul u e o he 20 h cen u y global e es ial
ca bon balance. Glob Chang Biol 13(3):679–706. h ps:// doi. o g/ 10. 1111/j. 1365- 2486. 2006. 01305.x
Bo cha d N, Schi mann M, Cayuela ML, Kammann C, W age-Mönnig N, Es a illo JM, Fue es-
Mendizábal T, Sigua G, Spokas K, Ippoli o JA, No ak J (2019) Biocha , soil and land-use in e ac ions
ha educe ni a e leaching and N2O emissions: a me a-analysis. Sci To al En i on 651:2354–2364.
h ps:// doi. o g/ 10. 1016/j. sci o en . 2018. 10. 060
Boysen LR, Luch W, Ge en D (2017) T ade-o s o ood p oduc ion, na u e conse a ion and clima e limi
he e es ial ca bon dioxide emo al po en ial. Glob Chang Biol 23(10):4303–4317. h ps:// doi. o g/
10. 1111/ gcb. 13745
Buss W, Wu ze C, Manning DAC, Rohling EJ, Bo e i z J, Mašek O (2022) Mine al-en iched biocha deli -
e s enhanced nu ien eco e y and ca bon dioxide emo al. Commun Ea h En i on 3(1):67. h ps://
doi. o g/ 10. 1038/ s43247- 022- 00394-w
Camps-A bes ain M, Amone e JE, Singh B, Wang T, Schmid HP (2015) A biocha classi ica ion sys em
and associa ed es me hods. In: Lehmann J, Joseph S (eds) Biocha o En i onmen al Managemen :
Science, Technology and Implemen a ion. Rou ledge, pp 165–193
Ca b ix. (2021). The wo ld’s la ges di ec ai cap u e and CO2 s o age plan is ON h ps:// www. ca b i x.
com/ he- wo lds- la ge s - di ec - ai - cap u e- and- co2- s o a ge- plan - is- on
Cha e jee S, Huang K-W (2020) Un ealis ic ene gy and ma e ials equi emen o di ec ai cap u e in deep
mi iga ion pa hways. Na Commun 11(1):3287. h ps:// doi. o g/ 10. 1038/ s41467- 020- 17203-7
Chiquie S, Pa izio P, Bui M, Sunny N, Mac Dowell N (2022) A compa a i e analysis o he e iciency,
iming and pe manence o CO2 emo al op ions. Ene gy En i on Sci 15(10):4389–4403
C ame W, Kickligh e DW, Bondeau A, Moo e B, Chu kina G, Nem y B, Ruimy A, Schloss AL, In e -
compa iso PPNM (1999) Compa ing global models o e es ial ne p ima y p oduc i i y (NPP):
o e iew and key esul s. Glob Chang Biol 5:1–15. h ps:// doi. o g/ 10. 1046/j. 1365- 2486. 1999.
00009.x
Die ich JP, Bodi sky BL, Humpenöde F, Weindl I, S e ano ić M, Ka s ens K, K eidenweis U, Wang
X, Mish a A, Klein D, Amb ósio G, A aujo E, Yalew AW, Baums a k L, Wi h S, Giannousakis A,
Beie F, Chen DMC, Lo ze-Campen H, Popp A (2019) MAgPIE 4 – a modula open-sou ce ame-
wo k o modeling global land sys ems. Geosci Model De 12(4):1299–1317. h ps:// doi. o g/ 10.
5194/ gmd- 12- 1299- 2019
