Hidden gems: small woody landscape features on agricultural land are overlooked in current assessments of agroforestry in Germany.

Vol.: (0123456789)
Ag o o es Sys (2025) 99:267
h ps://doi.o g/10.1007/s10457-025-01358-7
Hidden gems: small woody landscape ea u es
onag icul u al land a e o e looked incu en assessmen s
o ag o o es y inGe many
No aObladen · ZoeSchindle · Jona hanP.Sheppa d · Ka jaK öne · ElenaLa ysch ·
PawanDa a · ThomasSei e · Ch is ophe Mo ha
Recei ed: 11 Sep embe 2024 / Accep ed: 30 Sep embe 2025
© The Au ho (s) 2025
Abs ac In he ligh o escala ing clima e chal-
lenges, ag o o es y is ecei ing enewed a en ion
o i s po en ial o mi iga e g eenhouse gas emissions
while enhancing he esilience o ag icul u al sys-
ems. The main di e ence be ween ag o o es y and
con en ional ag icul u al sys ems is he p esence and
managemen o woody landscape ea u es (WLF).
The ew da ase s assessing WLF on ag icul u al land
a e limi ed in hei spa ial esolu ion and apply mini-
mum mapping h esholds, po en ially biasing de i ed
es ima es o WLF cha ac e is ics. Ou s udy aimed o
assess he cu en ex en o WLF in Ge many wi h
high spa ial esolu ion, including size a iabili y and
WLF ype composi ion. We in es iga ed WLF on
ag icul u al land ac oss se en ede al s a es. In each
o he se en s a es, 100 g id cells o alling 25 km2 o
ag icul u al land we e selec ed, amoun ing o a o al
a ea o 175 km2. Wi hin his a ea, WLF we e manu-
ally iden i ied, delinea ed and classi ied using digi al
o hopho os. The esul s we e compa ed wi h s a e-
o - he-a da ase s, pa icula ly he Digi al Basic
Landscape Model (ATKIS). We iden i ied a o al
o 4.8 km2 land hos ing WLF, co e ing 2.7% o he
in es iga ed a ea. The ex en o WLF es ima ed in ou
s udy was wice as la ge as he es ima e de i ed om
he ATKIS da ase . O e all, we iden i ied a much
la ge numbe o WLF, which we e on a e age much
smalle in size compa ed o he WLF in he ATKIS
Communica ed by Ge a do Mo eno.
No a Obladen and Zoe Schindle con ibu ed equally.
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/ s10457- 025- 01358-7.
N.Obladen· Z.Schindle · J.P.Sheppa d· K.K öne ·
E.La ysch· T.Sei e · C.Mo ha (*)
Chai o Fo es G ow h andDend oecology, Uni e si y
o F eibu g, F eibu g, Ge many
e-mail: ch is ophe [email p o ec ed]
N. Obladen
e-mail: [email p o ec ed]
Z. Schindle 
e-mail: [email p o ec ed]g.de
J. P. Sheppa d
e-mail: [email p o ec ed]
K. K öne 
e-mail: [email p o ec ed]g.de
E. La ysch
e-mail: elena.la ysc[email p o ec ed]
T. Sei e 
e-mail: homas.sei e [email p o ec ed]
P.Da a
Thü ingen Fo s AöR, Fo s liches Fo schungs- und
Kompe enzzen um (FFK), Go ha, Ge many
e-mail: pawanjee .da a@ o s . hue ingen.de
T.Sei e 
Depa men o Fo es andWood Science, S ellenbosch
Uni e si y, S ellenbosch, Sou hA ica
Ag o o es Sys (2025) 99:267
267 Page 2 o 21
Vol:. (1234567890)
da ase . Ex apola ing ou esul s o he na ional le el,
we es ima e WLF co e age a 4,899 km2, co espond-
ing o 57.2 Tg abo e-g ound biomass o 28.6 Tg
ca bon. The biomass and ca bon con en we e es i-
ma ed based on li e a u e-de i ed biomass densi ies.
Ou s udy indica es ha accu a ely assessing WLF
equi es highe spa ial esolu ions han p e iously
used. Da ase s based on low- esolu ion images and
hose using minimal mapping uni s a e no sui ed o
cap u ing small WLF.
Keywo ds T ees ou side o es s· Woody
pe ennials· Abo e-g ound biomass· Remo e
sensing· O hopho os· ATKIS
In oduc ion
In ecen yea s, ag o o es y as a dynamic and sus-
ainable land managemen p ac ice has gained
inc easing a en ion (Abbas e  al. 2017; Te asaki
Ha e al. 2023; Kuma e al. 2024; Jo anelly e al.
2025) o i s po en ial o add ess economic and eco-
logical challenges impac ing mode n ag icul u al
p ac ices posed by clima e change (He nández-Mo -
cillo e al. 2018). Simul aneously, in he con ex o
clima e change mi iga ion, he e has been g owing
in e es owa ds he assessmen o abo e-g ound bio-
mass (AGB) p oduc ion (Thapa e al. 2023) and he
inc eased oppo uni y o ca bon s o age (Nai 2012;
Schnell e al. 2015; Golicz e al. 2022; Sheppa d e al.
2024). Ag o o es y sys ems (AFS), as a o m o cli-
ma e-sma ag icul u e, exis in many empo al and
spa ial a angemen s bu a e p ima ily combina ions
o ees and sh ubs wi h li es ock o c opland (Nai
e al. 2021). Such woody s uc u es a e o en e e ed
o as woody landscape ea u es (WLF) and can be
conside ed an in eg al componen o AFS. In eg a -
ing WLF in ag icul u e no only enhances biodi e -
si y and ecosys em heal h bu also aligns wi h b oade
en i onmen al goals. The Eu opean G een Deal, o
example, seeks o es o e high-di e si y landscape
ea u es such as ee lines, ee g oups, and hedges on
a minimum o 10% o ag icul u al land by 2030 (DG
RTD 2021).
Al hough a ious ecological bene i s o AFS and
WLF a e widely ecognised (Jose 2009; Udawa a
e  al. 2019; Sollen-No lin e  al. 2020; Pan e a
e  al. 2021), he e is cu en ly a lack o de ailed,
quan i a i e knowledge abou AFS and WLF (Cas le
e al. 2022; Sa le e al. 2024; Scha e e al. 2024),
o ins ance ega ding hei spa ial ex en and dis-
ibu ion, as well as hei biomass (Schnell e  al.
2015; Golicz e al. 2022). The p o ision o baseline
es ima es ega ding he cu en s a us o WLF could
yield aluable in o ma ion ha could be applied o
mul iple pu poses. Fo example, es ima es o he
AGB and ca bon con en o WLF could be used o
es ima ing hei clima e change mi iga ion po en ial.
Such in o ma ion is ele an o ecological analyses,
bu also as an aid o poli ical decisions. To assess
he ca bon con en o o he cha ac e is ics o WLF,
in o ma ion on he ex en o he WLF is equi ed.
And he mo e accu a e he inpu da a, he be e
he de i ed es ima es o ca bon con en and WLF
cha ac e is ics.
Despi e la ge echnological ad ances in geospa-
ial da a collec ion and emo e sensing o e he las
decades, eliable in o ma ion abou he ex en o
WLF on ag icul u al land on na ional scales emains
sca ce. Cu en ly he mos p e alen land co e and
land use da ase s (LULC) a ailable a he Eu opean
le el which con ain in o ma ion conce ning WLF a e
he Land Use/Co e A ea ame Su ey (LUCAS),
he CORINE land co e (CLC), he CLC + Back-
bone (CLC +), and he High Resolu ion Laye Small
Woody Fea u es (HRL-SWF) da ase s (Table 1).
LUCAS is a da a collec ion based on upscaled poin
samples and ield obse a ions. The CLC, CLC + and
HRL-SWF, on he o he hand, a e based on sa elli e
da a. On a Ge man na ional le el, he Digi al basic
landscape model (Basis-DLM), p o ides a ious
de ailed geospa ial in o ma ion and includes in o ma-
ion on land co e and land use (AdV 2024b). The
Basis-DLM is a high- esolu ion da ase which is pa
o he Au ho i a i e Topog aphic Ca og aphic In o -
ma ion Sys em (ATKIS). I is based on opog aphic
maps, digi al o hopho os (DOP) and some addi ional
da a, and is upda ed egula ly (Jäge 2003). The da a-
se is compiled om he da a collec ed by indi idual
ede al s a es. Due o di e ing sampling app oaches,
he spa ial esolu ion and he ime o da a collec ion
a e no uni o m ac oss all s a es. While he e may be
o he land use da ase s a ailable o speci ic egions
wi hin Ge many, he Basis-DLM is one o he mos
p ominen and widely used da ase s ha includes bo h
opog aphic and land use in o ma ion a he Ge man
na ional scale.
Ag o o es Sys (2025) 99:267 Page 3 o 21 267
Vol.: (0123456789)
Such LULC da ase s p esen a ious limi a ions
conce ning ei he hei opicali y, accessibili y and/
o spa ial esolu ion. Fo ins ance, LUCAS does no
spa ially map WLF bu ins ead p o ides a ea es i-
ma es o WLF p esence and ypes ac oss Eu ope
based on ield-sampled poin s (Buck e  al. 2015;
Rubio-Delgado e al. 2024). Mos o he da ase s apply
a ied minimum and maximum h esholds o map-
ping uni s, and some o he wid hs and leng hs o
linea s uc u es (see Table1). While LULC da ase s
a e no designed o ep esen biomass di ec ly, hei
spa ial limi a ions can esul in he sys ema ic omis-
sion o small-scale landscape ea u es such as WLF.
As a esul , he ex en o WLF, including he biomass
and ecological unc ions hey con ibu e o, may be
unde es ima ed when such da ase s a e used o appli-
ca ions such as land-use planning o ca bon accoun -
ing, also in he con ex o AFS.
Ea lie li e a u e has es ima ed he ex en o ei he
AFS o WLF on ag icul u al land in Eu ope and
Ge many based on hese da ase s. Howe e , hese
es ima es di e conside ably be ween s udies, likely
explained by he di e en cha ac e is ics and limi a-
ions o he unde lying da a as desc ibed abo e (see
Table1). Fo example, den He de e al. (2017) es i-
ma ed ha 8.8% o he ag icul u al a ea in he EU is
unde AFS managemen , wi h he es ima e o Ge -
many amoun ing o 1.6% o he o al ag icul u al
a ea. Acco ding o Golicz e al. (2021), he cu en
ex en o WLF in Ge many o als 4.6% o he ag i-
cul u al land. Using an upda ed e sion o he da a-
se , he EEA (2024c) de e mined a compa able ex en
o 4.8%. E en hough he Basis-DLM is one o he
mos used land co e da ase s wi hin Ge many, is
publicly a ailable, and has a highe spa ial esolu ion
han mos sa elli e da ase s, we a e no awa e o any
s udies o da e using his da a o es ima e he ex en o
WLF in Ge many.
The objec i e o his s udy was o assess he cu -
en ex en o WLF on ag icul u al land in Ge many.
Speci ically, we analysed he o al WLF a ea as well
as he a iabili y in size and sha es o di e en WLF
ypes ac oss se en ede al s a es. This was accom-
plished by u ilising high- esolu ion o hopho os wi h
a spa ial esolu ion o 20cm o a manual delinea-
ion and classi ica ion o WLF ea u es in sample g id
cells. An es ima ion o WLF ex en on hese same
sample g id cells based on he Basis-DLM allowed
Table 1 O e iew on land use and land co e da ase s con-
aining in o ma ion on woody landscape ea u es. The mini-
mum and maximum mapping sizes o he ele an ea u es
o each da ase a e gi en as minimum/maximum mapping
a ea (MMA), minimum/maximum mapping wid h (MMW),
and minimum/maximum mapping leng h (MML). Fo as e -
based da ase s, he MMA co esponds o he g ound a ea co -
e ed by one single pixel. The ull names o he da ase s a e:
Land Use/Co e A ea ame Su ey (LUCAS), CORINE land
co e (CLC), CORINE land co e plus Backbone (CLC+),
High Resolu ion Laye –Small Woody Fea u es (HRL-SWF),
Digi al basic landscape model (Basis-DLM). The da ase s we e
used o AFS o WLF es ima ion by 1den He de e al. (2017),
2Golicz e al. (2021) and 3 he EEA (2024c)
Da ase ( elease) Re e ences Co e age Sampling MMA MMW MML
LUCAS
(2012)1Buck e al. (2015) Eu ope Sample poin s
wi h ield
da a
≥ 2,500 m2 o ≥ 10,000 m2
n.a n.a
CLC
(2018) Bü ne e al. (2017) Eu ope Sa elli e da a ≥ 250,000 m2 ≥ 100m n.a
CLC + ec o (2018) EEA (2022, 2024b) Eu ope Sa elli e da a ≥ 5,000 m2 ≥ 20m n.a
CLC + as e (2021) EEA (2022, 2024a) Eu ope Sa elli e da a ≥ 100 m2n.a n.a
HRL-SWF
(2015)2EEA (2020) Eu ope Sa elli e da a ≥ 200 m2 ≤ 5,000 m2 ≤ 30m ≥ 50m
HRL-SWF
(2018)3CLMS (2018) Eu ope Sa elli e da a ≥ 200 m2 ≤ 5,000 m2 ≤ 30m ≥ 30m
Basis-DLM (n.d.) AdV (2021, 2024b) Ge many Topog aphic
maps, digi al
o hopho os
≥ 1,000 m2 o ≥ 5,000 m2
o ≥ 10,000 m2n.a ≥ 200m
Ag o o es Sys (2025) 99:267
267 Page 4 o 21
Vol:. (1234567890)
o compa isons be ween digi al o hopho os and he
Basis-DLM a an indi idual ede al s a e le el. Addi-
ionally, we aimed o an upscaling o WLF a ea om
sampling a eas o ede al s a e and an ex apola ion o
na ional le els alongside an es ima ion o AGB and
ca bon s o age wi hin WLF on ag icul u al land in
Ge many.
Ma e ial & me hods
Resea ch a ea
The s udy was ca ied ou wi hin se en ede al
s a es o Ge many (Fig. 1): Lowe Saxony (NI),
No h Rhine-Wes phalia (NW), B andenbu g (BB),
Hesse (HE), Saxony-Anhal (ST), Saxony (SN), and
Thu ingia (TH). All o he s a es we e excluded, ei he
due o una ailable LULC da a o insigni icance due
o a low ex en o ag icul u al a ea, i.e. ci y s a es.
An o e iew o e he se en selec ed s a es and hei
sha es o land unde ag icul u al use a e shown in
Fig.1.
The wes o Ge many gene ally has a empe a e
oceanic clima e (C b), while highe ele a ions in he
sou h and eas e n egions expe ience a wa m-summe
humid con inen al clima e (D b), wi h localized sub-
a c ic condi ions (D c) in alpine a eas (Beck e  al.
2018). Ge many has a mean annual p ecipi a ion sum
o 729mm and an a e age mean su ace ai empe a-
u e o 9.6°C (Wo ld Bank 2011). The se en sampled
ede al s a es expe ience a highe a e age mean su -
ace ai empe a u e o 10 °C wi h li le a iabili y
be ween s a es, while he mean annual p ecipi a ion
Fig. 1 a Map o Ge many illus a ing he selec ed (ligh
g een) and no selec ed s a es (da k g een). b De ailed o e -
iew o he selec ed s a es included in his s udy. The doughnu
plo s depic he sha e o ag icul u al land (da k o ange) and he
sha e o land unde o he land uses (ligh o ange) (Map da a:
Bundesam ü Ka og aphie und Geodäsie (2021), Geo da a:
Es i (2024), G aphics da a: S a is isches Bundesam (2024))
Ag o o es Sys (2025) 99:267 Page 5 o 21 267
Vol.: (0123456789)
sum o 678 mm is lowe han he na ional a e age
and shows dec easing p ecipi a ion on a wes o eas
g adien .
Da a sou ces
To de ine he se en selec ed s a es, s anda dised
adminis a i e bo de s (Bundesam ü Ka og a-
phie und Geodäsie 2021) we e used. Addi ional da a
we e sou ced om he ATKIS da abase, depending
on open-access a ailabili y, which a ied be ween
ede al s a es. Speci ically, we used he Basis-DLM
( e e ed o as he ATKIS-DLM he ea e ) as digi al
landscape model da a, alongside digi al o hopho os
(DOPs). The ATKIS-DLM desc ibes opog aphical
objec s ega ding hei spa ial posi ion and a ange-
men , ype and desc ip i e a ibu es, i.e. addi ional
hema ic o quali a i e in o ma ion (AdV 2024b).
DOPs a e dis o ion- ee, ue- o-scale pho og aphic
images o he Ea h’s su ace which a e de i ed om
ae ial pho og aphs. In he ATKIS da a, hey a e a ail-
able a di e en g ound esolu ions (AdV 2024a). Fo
ou pu pose, we used DOPs wi h a g ound esolu ion
o 20cm. The unde lying ae ial pho os we e acqui ed
be ween 2019 and 2023, depending on he s a e
and he iles’ opicali y. Addi ionally, Google Maps
image y (Google 2023) was used o occasionally ali-
da e he esul s o o ob ain a di e en seasonal pic-
u e o he landscape.
S udy design & da a collec ion
A g id wi h a cell size o 500m × 500m (0.25 km2)
was o e laid on each ede al s a e (Fig.2a, Fig.3a).
Wi hin he AdV objec ca alogue, he di e en ypes
o opog aphic objec s a e desc ibed in de ail (AdV
Fig. 2 Visualisa ion o he s udy wo k low. a Sampling o he
g id cells wi hin he selec ed ede al s a es. b Ex ac ion o
ATKIS WLF in he selec ed g id cells. c Manual delinea ion
and classi ica ion (MDC) o he WLF in he selec ed g id cells.
d Es ima ing he o al WLF a ea pe selec ed ede al s a e om
he ATKIS-WLF and he MDC-WLF. e Es ima ing he o al
WLF biomass pe selec ed ede al s a e om he MDC WLF

Ag o o es Sys (2025) 99:267
267 Page 6 o 21
Vol:. (1234567890)
2021). In e ms o opicali y, some objec s a e upda ed
annually (peak opicali y, e.g. a ic). To in es iga e
only ag icul u al a eas, all g id cells in e sec ing wi h
opog aphic objec ypes designa ed as u ban a ea
(code 41,000), o es (code 43,002) and wa e bodies
(code 44,000) we e excluded om he g id, and hus,
also om he analysis (Fig.2a, Fig.3b). Addi ionally,
g id cells in e sec ing wi h objec s o he alue ype
s ee s (code 42,002) unde he b oade objec ype
a ic (code 42,000) we e excluded o elimina e la ge,
sealed a eas om ou su ey. F om he emaining
g id, 100 g id cells pe s a e we e andomly selec ed
o u he analysis (Fig.2a, Fig.3c), esul ing in 700
g id cells in o al. This esul ed in a o al s udy a ea
o 25 km2 pe s a e, amoun ing o a o al s udy a ea o
175 km2.
Wi hin he selec ed g id cells, WLF we e iden i-
ied and mapped manually on DOPs as ec o poly-
gons (Fig. 2c). The WLF we e isually classi ied
in o se en di e en ea u e ypes, namely hedge ow,
hedge ow wi h gaps, ee ow, g o e, sh ub, ee and
o cha d (Table2, Fig.4). In he ollowing, he da a
ob ained wi hin his s udy is e e ed o as “manual
delinea ion and classi ica ion” (MDC).
In o de o compa e he mapped WLF om he
MDC o he ATKIS-DLM da ase , he objec ypes
Fig. 3 a Example o he ull g id be o e exclusions, b a e exclusion o u ban a ea, o es , wa e bodies and ede al highways and c
one andomly selec ed g id cell as used o ou s udy (Map da a: Es i (2024))
Table 2 De ini ions o he woody landscape ea u e (WLF) ypes used in he manual delinea ion and classi ica ion
WLF ype De ini ion
Hedge ow Sh ubs wi h o wi hou ees a anged in ows
Hedge ow wi h gaps Sh ubs wi h o wi hou ees a anged in ows wi h small gaps in-be ween
T ee ow T ees plan ed in ows wi h < 10m dis ance o each o he
G o e T ee g oups wi h > 2 indi iduals, oge he wi h sh ubs bu domina ed by ees
Sh ub Sh ubs wi h o wi hou single ees in-be ween which a e no a anged in ows
T ee Indi idual ees
O cha d Collec ion o ees managed o e.g. ui o nu p oduc ion, wi h egula o
i egula spacing and a ied managemen in ensi y
Ag o o es Sys (2025) 99:267 Page 7 o 21 267
Vol.: (0123456789)
woody ea u es (code 43,003) and ege a ion ea-
u e (code 54,001) we e agg ega ed o desc ibe he
WLF wi hin he ATKIS da ase (Fig. 2b). In he
ollowing, his agg ega ed da a will be e e ed o
as ATKIS-WLF. In he ATKIS-DLM, sca e ed
o cha ds and o cha ds a e lis ed unde ag icul u e
(code 43,001). Since hese alue ypes we e no
accessible o all se en selec ed ede al s a es, hey
we e no included in he ATKIS-WLF.
Fig. 4 Examples o he se en woody landscape ea u e ypes ob ained in he manual delinea ion and classi ica ion. The images a e
illus a ing he ypes a hedge ow, b hedge ow wi h gaps, c ee ow, d g o e, e sh ub, ee and g o cha d
Ag o o es Sys (2025) 99:267
267 Page 8 o 21
Vol:. (1234567890)
A ea and biomass es ima ion
We upscaled bo h o al and ype-speci ic WLF a eas o
he ede al s a e le el by compu ing hei sha es in he
sampled ag icul u al a ea and scaling hem using he
o al ag icul u al a ea o each s a e (see Fig.2d). The
d y woody AGB wi hin he di e en ede al s a es and
WLF ca ego ies was es ima ed by applying biomass
densi ies epo ed in p e ious s udies o he a ea o
he espec i e WLF ca ego ies (see Table3, Fig.2e).
The AGB o he ca ego ies hedge ow, hedge ow wi h
gaps, ee ows, g o es and sh ub we e es ima ed using
insigh s om G een e al. (2021).
In o de o calcula e he AGB o indi idual ees,
open g own walnu ees (Juglans egia L.; a common
species employed wi hin AFS, c. . Reisne e al. (2007)
and Pa don e al. (2020)) we e used as a ep esen a i e
ee species (Table3). The ee biomass was de i ed by
i s es ima ing he diame e a b eas heigh (DBH, in
cm) om he c own p ojec ion a ea (CPA, in m2) using
a ea anged equa ion gi en by Schindle e al. (2023b)
(Eq.1). This s ep was applied since he e was no unc-
ion a ailable which di ec ly in e ed AGB om CPA.
Using he es ima ed DBH (in cm), he AGB (in kg) was
calcula ed o each ee using he co esponding equa-
ion by Schindle e al. (2023b) (Eq.2). Fo example, a
ee wi h a CPA o 71.5 m2 and a co esponding diam-
e e o 30cm would ep esen 745kg AGB.
(1)
DBH =e(
ln
(
CPA
1.0638
)
+2.038
)
∕
1.819
(2)
AGB =e−1.8324+2.4675∗ln(DBH)∗1.0542
Fo he ca ego y o cha d, a densi y o 100
ees pe hec a e wi h a mean DBH o 20cm was
assumed. In Ge many, i is common o plan ees
in o cha ds wi h a ee spacing o 10m (e.g. Kom-
pe enzzen um Ökolandbau Niede sachsen 2019).
AGB was calcula ed d awing on he insigh s o
Schindle e al (2023a) ega ding che y ees (P u-
nus a ium L.). This me hodology ensu es a nuanced
and con ex -speci ic app oach o AGB es ima ion
ac oss di e se landscape ea u es u ilising a ep e-
sen a i e ee species ha has wide applicabili y.
To p o ide an ou look, we es ima ed d y woody
AGB on ag icul u al land wi hin WLF a he
na ional le el by ex apola ing es ima es om he
ede al s a e le el up o he na ional le el. This
ex apola ion elies on he assump ion ha he ela-
i e ex en o WLF wi hin ag icul u al land ac oss
Ge many mi o s ha o he sampled ede al s a es,
wi h he p opo ions o WLF ypes app oxima ed
based on he composi ion o WLF ypes in he se en
analysed s a es. The ca bon s o ed in AGB was es i-
ma ed assuming a 50% con en in d y woody bio-
mass (Thomas and Ma in 2012). Ca bon con en
alues we e con e ed o CO2 equi alen s (CO2e)
by mul iplying hem by 3.67, which co esponds o
he di e ence in a omic weigh (Gues e al. 2013).
The da a p epa a ion, collec ion as well as geo-
spa ial p ocessing (Fig. 2a,b,c) we e ca ied ou
wi h he open-sou ce so wa e QGIS e sion 3.28.1
(QGIS De elopmen Team 2022). Da a analyses
(Fig.2d,e) we e conduc ed using he s a is ics so -
wa e R e sion 4.5.1(R Co e Team 2025).
Table 3 O e iew o he used d y abo e-g ound biomass
(AGB) pe a ea o each woody landscape ea u e (WLF) ype.
The e e ence column e e s o he e e ence om which he
es ima e was ob ained. Fo each ca ego y excep o ees, a
single alue was applied o he WLF a ea. Fo ees, he bio-
mass was de i ed based on he c own p ojec ion a ea o indi-
idual ees
WLF ype AGB (Mg km−2) Re e ences
Hedge ow 9,900 Unmanaged hedge ows (G een e al. 2021)
Hedge ows wi h gaps 5,560 Managed hedge ows (G een e al. 2021)
T ee ows 24,800 T ee lines (G een e al. 2021)
G o es 14,460 Woodland (G een e al. 2021)
Sh ub 100 Sc ub (G een e al. 2021)
T ee – Biomass unc ion o Juglans egia (Schindle e al. 2023b)
O cha d 1,860 Biomass unc ion o P unus a ium based on 100 ees
pe hec a e each wi h a DBH o 20cm (Schindle e al.
2023a)
Ag o o es Sys (2025) 99:267 Page 9 o 21 267
Vol.: (0123456789)
Resul s
WLF de ec ion & a ea
As a esul o he MDC me hodology, 88.6% o he
in es iga ed g id cells we e ound o con ain one o
mo e WLF (Fig.5a). Wi hin hese g id cells con ain-
ing a leas one WLF, 11,136 ea u es we e de ec ed,
co e ing a o al a ea o 4.8 km2 cons i u ing 2.7% o
he in es iga ed a ea (Fig.5b). Upscaling ou esul s
o he o al ag icul u al a ea o he se en selec ed
s a es yields a sum o 2,640 km2 WLF a ea (Appendix
Table4). Bo h he o al ex en and he ela i e ex en
o WLF wi hin ag icul u al land designa ion (WLF%)
a ies conside ably be ween s a es (Fig.5, Appendix
Table4, Appendix Fig.8). Fo ins ance, he WLF%
in HE was mo e han wice as la ge as in BB, NW,
ST, o SN. As he o al ag icul u al a ea in Ge many
amoun s o a ound 180,207 km2 (S a is isches Bunde-
sam 2024), we es ima e WLF in Ge many o co e
an a ea o 4,899 km2 when assuming a simila WLF
dis ibu ion o all ede al s a es in Ge many.
Using he ATKIS-DLM, on he o he hand, WLF
we e de ec ed only in 26.9% o he g id cells. Wi hin
hese, 312 ea u es (2.8% o he MDC) we e de ec ed,
co e ing a o al a ea o 2.1 km2 (44.5% o he MDC),
o alling 1.2% o he in es iga ed a ea. The la ge di -
e ences in he numbe and o al a ea o WLF be ween
he MDC and he da a de i ed om he ATKIS-DLM
can be obse ed in all in es iga ed s a es (Fig.5).
The sizes o indi idual WLF in he MDC anged
om 0.44 m2 o 122,560 m2 (Fig. 6a). The o e -
all in e qua ile ange (IQR) spanned om 24 m2 o
139 m2. In compa ison, he sizes o he WLF iden-
i ied om he ATKIS-WLF anged be ween 0.04 m2
and 177,629 m2 (Fig.6b). The IQR o he ATKIS-
WLF da ase anged be ween 1,297 m2 and 6,862 m2.
The a e age size o he WLF iden i ied in he MDC
di e ed signi ican ly be ween he s a es (Fig.6a). In
he case o he ATKIS-WLF, we ound no signi ican
di e ences be ween s a es (Fig.6b). A compa ison o
he dis ibu ion o WLF sizes shows ha he ATKIS-
WLF we e gene ally la ge han he WLF iden i ied
in he MDC (Fig.6c).
WLF ype composi ion & ca bon s o age
Ac oss all in es iga ed s a es, a o al o 52,942 Mg
o d y woody AGB was es ima ed o he su eyed
a ea wi hin he g id cells, whe eby HE, NI and TH
show he highes amoun s (Fig.7a). The p opo ions
o he di e en WLF ypes a y conside ably be ween
s a es, e.g., HE displays a high sha e o hedge ows
and g o es, whe eas NW shows he la ges ee ow
a ea among he se en in es iga ed s a es. In BB,
hedge ows accoun ed o mo e han hal o he WLF
a ea, while in NI only a qua e o he WLF could be
a ibu ed o hedge ow s uc u es.
This densi y o AGB, i.e. he AGB pe ag icul-
u al a ea, a ied be ween s a es and inc eased in he
o de o SN < BB < ST < NW < TH < NI < HE. Again,
he di e ences we e conside able, wi h HE s o ing
app oxima ely wice as much AGB pe ag icul u al
a ea compa ed o SN, BB and ST. When conside ing
he sha es o a ea and WLF ype composi ion, hedge-
ows ha e he highes a ea sha e ollowed by g o es,
whe eas g o es ha e he highes AGB sha e ollowed
by hedge ows (Fig.7b). T ee ows, on he o he hand,
occupy only a small a ea while accoun ing o a high
p opo ion o biomass. An ex apola ion o he o al
a ea o each s a e shows ha he s a es di e ed con-
side ably in hei WLF biomass con en . Fo example,
NI boas s signi ican ly mo e WLF biomass in com-
pa ison wi h he o he s a es (Fig.7c). Con e sely, he
quan i ies o biomass in NW and HE a e ela i ely
simila , al hough hey a e composed o di e en WLF
ypes. Summing up hese ex apola ions, he se en
in es iga ed s a es con ain in o al 30.8Tg o AGB,
which is equi alen o 15.4 Tg ca bon o 56.6 Tg
CO2e.
Unde he assump ion ha he selec ed ede al
s a es a e ep esen a i e o Ge many as a whole, in
e ms o hei WLF ex en and ype composi ion (see
Fig.7b), an ex apola ed na ional es ima ion o AGB
was calcula ed. In o al, we es ima e ha Ge many
con ains 57.2Tg o AGB wi hin WLF. This is equi a-
len o 28.6Tg ca bon o 105.0Tg CO2e. Mo eo e ,
we can sugges ha a p esen WLF co e s a land a ea
o 4,899 km2 in Ge many.
Discussion
This s udy aimed o e alua e whe he applying
MDC me hods o WLF in Ge many p oduces di -
e en es ima es o WLF ex en and biomass com-
pa ed o exis ing da a a ailable in he ATKIS-
DLM. WLF iden i ied wi hin he ATKIS-DLM
Ag o o es Sys (2025) 99:267
267 Page 16 o 21
Vol:. (1234567890)
o d y woody AGB may be held wi hin WLF in Ge -
many. This equa es o 28.6 Tg ca bon o 105.0 Tg
CO2e, which is a signi ican amoun in he con ex o
ca bon budge ing. Since he e a e e en ewe publica-
ions on he ex en han on he biomass and ca bon o
WLF, he e is li le oppo uni y o compa e ou es i-
ma e wi h hose o o he s udies. O he p e iously
men ioned s udies, only Golicz e al. (2021) p o ided
a co esponding es ima e (36.3Tg ca bon). Howe e ,
as hey included bo h abo e- and belowg ound bio-
mass in hei analysis, hei esul s a e no compa a-
ble wi h ou s.
I should be emphasised ha ou es ima e was
de i ed unde a se o assump ions. Speci ically, he
limi ed numbe o ede al s a es analysed and he use
o biomass densi ies om o he geog aphic egions
(e.g. G een e al. (2021) om I eland and Schindle
(2023a; b) om he Rhine Valley in Sou he n Ge -
many) should be c i ically conside ed. Using spe-
ci ic biomass densi ies om he sampled egions o
egions wi h simila g owing condi ions (e.g. cli-
ma ic condi ions and managemen ), could imp o e
he accu acy o AGB es ima es. Mo eo e , applying a
single biomass densi y pe WLF ype is a s ong sim-
pli ica ion. As o example, he AGB in hedge ows
a ies conside ably depending on hedge ow s uc u e,
species composi ion, heigh and managemen (Axe
e al. 2017; Black e al. 2023).
Example scena io
E iden ly. a key limi a ion in exis ing es ima es o
WLF ex en and ca bon es ima es based on sa elli e
image y is he p e iously men ioned unde ep esen-
a ion o small WLF due o limi ed spa ial esolu ion
and minimum mapping uni s. To highligh he ele-
ance o hese o en-o e looked ea u es, we p esen
an example scena io based on he allome ies p e-
sen ed by Schindle e al. (2023b), which we e used in
his s udy o es ima e ee AGB. Assuming ha e e y
hec a e o ag icul u al land in Ge many would ha e
an addi ional i e walnu ees wi h a unk ci cum e -
ence o 20cm each, his would esul in an addi ional
3,082 km2 o WLF, i.e. 1.7% o he ag icul u al a ea
in Ge many. In e ms o biomass, his would co e-
spond o 24.7Tg AGB (i.e. 12.3Tg ca bon, 45.3Tg
CO2e). As walnu ees o his diame e ha e a c own
wid h o abou 6m, hey would likely go unde ec ed
in da ase s wi h a low spa ial esolu ion, unless hey
we e g ouped oge he . This example illus a es how
he cumula i e impac o many small WLF can sig-
ni ican ly in luence es ima es a a la ge scale. Thus,
while ou app oach used highe esolu ion da a, he e
is a con inuing need o be e esol e small ea u es in
ca bon budge ing.
Me hodological imp o emen s & ou look
Using he p esen ed me hodology, we we e able o
es ima e he ex en , biomass and ca bon con en o
WLF in Ge many a bo h s a e and na ional le el. We
demons a ed he impo ance o using high- esolu ion
image y o de ec small WLF ha would o he wise be
o e looked. Ne e heless, ou esul s a e subjec o
a ious unce ain ies. In he ollowing, we ecognise
he limi a ions o he applied me hodology and make
sugges ions o u he imp o emen s.
Rega ding sampling me hod, he numbe o ana-
lysed g id cells is he main sou ce o unce ain y. To
ensu e compa abili y be ween s a es a he g id cell
le el, he same numbe o g id cells was sampled in
each s a e, ega dless o he o al ag icul u al a ea o
he espec i e s a es. Howe e , his means ha ou
esul s a e mo e ep esen a i e o smalle s a es han
o la ge s a es, as a la ge p opo ion o he o al
ag icul u al a ea is sampled. In la ge s a es, we a e
mo e likely o ha e andomly unde ep esen ed less
common ypes o ag icul u e, e.g. ag icul u al land
ha is cul i a ed wi h low in ensi y. Ano he sou ce
o unce ain y is he ex apola ion o he esul s om
he selec ed s a es o he whole o Ge many, because
he ela i e ex en and ype composi ion o WLF a -
ies among ag oecological egions and his is likely o
be e lec ed in di e ences among ede al s a es. An
inclusion o all s a es in he analysis would imp o e
he eliabili y o ou esul s. While da a a ailabili y
p e en ed a ull na ional analysis, we assume ha
he su eyed s a es a e somewha ep esen a i e o
Ge many. The se en analysed ede al s a es, co e -
ing only abou hal o Ge many’s land a ea, cap u ed
he wes –eas g adien be ween o me Wes and Eas
Ge many, which we assume has a la ge in luence on
WLF dis ibu ion.
The MDC me hodology may also ha e in o-
duced some unce ain y in he esul s. Al hough we
used e y high- esolu ion da a compa ed o p e i-
ous da ase s, he manual delinea ion and classi ica-
ion was no unambiguous o all WLF. Depending

Ag o o es Sys (2025) 99:267 Page 17 o 21 267
Vol.: (0123456789)
on he ea u e size and he DOPs, some WLFs we e
mo e di icul o delinea e and classi y han o he s.
Fo example, he iming o he da a collec ion o
he DOPs had a majo in luence due o he seasonal
in luence on he oliage and he in luence o he
sun’s posi ion on shade cas . Al hough humans a e
e y good a ecognising pa e ns (Pa ikh and Zi -
nick 2010), any MDC is likely subjec i e o some
deg ee. In his s udy, bo h delinea ion and classi i-
ca ion o WLF we e likely biased o some ex en by
his subjec i i y. The e o e, o minimise inconsis -
encies, all MDC was conduc ed by a single pe son.
To u he imp o e he eliabili y o ou esul s,
o hopho os wi h a highe spa ial esolu ion could
be used o acili a e he MDC o WLF. Howe e ,
such da a we e nei he a ailable no a o dable
in he con ex o his s udy. In addi ion, he inclu-
sion o 3D da a as opposed o 2D da a could aid
ea u e classi ica ion and imp o e he accu acy
o AGB es ima es (Lingne e  al. 2018). Finally,
AGB es ima es could be imp o ed by inco po a ing
sh ub and ee species iden i y in o he calcula ion.
Al hough au oma ed app oaches o classi ying ee
species om emo e sensing da a a e al eady a ail-
able, hese app oaches ypically use da a a a much
highe spa ial esolu ion han was a ailable (Egli
and Höpke 2020; Schie e e al. 2020). In conclu-
sion, once da a wi h a highe esolu ion o dimen-
sion a e a ailable, AGB es ima es can be imp o ed.
As an ou look, we p opose ha u u e esea ch
should ocus on he de elopmen and op imisa ion
o au oma ic classi ica ion o WLFs o imp o e
es ima es o WLF ex en . Al hough we a e posi i e
ha ou app oach is a iable op ion o assessing
he ex en o WLFs on small scales, i is no p ac-
ical o de ec ing WLFs on la ge scales. Once
la ge a eas can be au oma ically classi ied, he
ag icul u al a ea could be ully classi ied, and hus,
egional di e ences could be be e cap u ed han
wi h a sampling app oach. Fo au oma ic classi i-
ca ion, deep lea ning app oaches could be used. As
summa ised by Ch is in e al. (2019), deep lea ning
has ecen ly e olu ionised a ious esea ch a eas
and can be a powe ul ool o esea che s. How-
e e , aining deep lea ning algo i hms equi es
high compu ing powe and la ge amoun s o ain-
ing da a. In he u u e, a leas he challenge o
compu a ionally in ensi e aining is likely o be
educed hanks o echnological p og ess.
Conclusion
Al hough he nume ous po en ial bene i s o WLFs
in he ag icul u al landscape a e well known, he e
is li le de ailed, high- esolu ion da a on he ex en o
hese s uc u es. Exis ing da ase s a e mos ly based on
low- esolu ion sa elli e da a and apply minimum map-
ping uni s, which can lead o small-scale WLF being
o e looked in hese su eys. In his s udy, we mapped
WLF in Ge many using high- esolu ion o hopho-
os and compa ed hem wi h an es ablished da ase .
Compa ed o he ATKIS da ase , we iden i ied con-
side ably mo e, and also, conside ably smalle WLFs,
in sum co e ing an a ea abou wice as la ge as he
ex en es ima ed in he ATKIS da ase . These di e -
ences emphasise ha he use o high- esolu ion da a
is necessa y o mo e accu a e es ima ions o WLF
ex en , because a la ge numbe o small elemen s can
ha e subs an ial e ec s in agg ega e. Ou s udy also
indica es ha he s a us quo o he WLF dis ibu ion
in Ge many, and p obably also in o he coun ies,
may be unde es ima ed. The insigh s we ha e gained
a e pa icula ly ele an o accu a e ca bon budge -
ing, which is cu en ly o high ele ance in he con-
ex o clima e change. They can also be use ul o
policy make s in ol ed in he p omo ion o AFS. Fo
unding guidelines o be e ec i e, WLF dis ibu ion
and ca bon budge es ima es canno be disconnec ed
om eali y. An accu a e assessmen o he cu en
s a us could possibly help o p e en such a discon-
nec ion. Fu u e s udies on WLFs should, o his ea-
son, aim o include small WLFs. Fu he mo e, u u e
esea ch could include addi ional WLF cha ac e is ics
o allow o a mo e accu a e assessmen o ca bon
s o age in hese s uc u es. Including small WLF will
make accoun ing o he cu en a ea and olume o
woody componen s on ag icul u al land mo e accu-
a e. This would allow a mo e ealis ic p edic ion o
he de elopmen o long- e m ca bon s ocks in ees
and hedges ou side o es s, leading o a mo e ealis-
ic ime ame o achie ing he EU G een Deal a -
ge o ensu ing ha a leas 10% o ag icul u al land
is co e ed by woody ege a ion by 2030. Mo eo e ,
in-si u o 3D da a could be used o inco po a e he
i ali y and in e nal s uc u e o WLF. Such in o ma-
ion would no only be ele an o a mo e accu a e
assessmen o ca bon s o age bu could also be used
o assess o he e ec s o WLFs, such as he enhance-
men o biodi e si y o socio-economic bene i s.
Ag o o es Sys (2025) 99:267
267 Page 18 o 21
Vol:. (1234567890)
Acknowledgemen s We would like o hank Janusch Jehle
o his suppo and his ideas ega ding he me hodology.
Au ho s’ con ibu ions NO and ZS con ibu ed equally.
Concep ualiza ion: NO, CM; Me hodology: NO, CM, PD;
In es iga ion: NO; Fo mal analysis: NO, ZS; Visualiza ion:
NO; W i ing—o iginal d a p epa a ion: NO, ZS, JS, KK, EL;
W i ing— e iew and edi ing: NO, ZS, KK, JS, EL, PD, CM,
TS; Funding acquisi ion: CM, TS; Supe ision: CM.
Funding Open Access unding enabled and o ganized by
P ojek DEAL. This s udy is unded by he ollowing p ojec s:
The p ojec INTEGRA, g an numbe 2819NA071, is sup-
po ed by unds o he Ge man Fede al Minis y o Food and
Ag icul u e (BMEL) based on a decision o he pa liamen o
he Fede al Republic o Ge many ia he Fede al O ice o
Ag icul u e and Food (BLE) unde he Fede al P og amme
o Ecological Fa ming and O he Fo ms o Sus ainable Ag i-
cul u e. The MODEMA p ojec , g an numbe 2222NR061J,
is suppo ed by unds o he Fede al Minis y o Ag icul u e,
Food and Regional Iden i y (BMLEH) based on a decision o
he Pa liamen o he Fede al Republic o Ge many ia Agency
o Renewable Resou ces (FNR) unde he unding p og amme
“Sus ainable Renewable Resou ces”. The p ojec MARVIC is
co- unded by he Eu opean Union, g an numbe 101112942.
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. Nei he he Eu opean Union no he g an ing au ho i y
can be held esponsible o hem.
Da a a ailabili y All da a suppo ing he indings o his
s udy a e a ailable wi hin he pape and i s Supplemen a y
In o ma ion.
Decla a ions
Con lic o in e es s The au ho s decla e no compe ing in e -
es s.
Open Access This a icle is licensed unde a C ea i e Com-
mons 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 Commons 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/.
Re e ences
Abbas F, Hammad HM, Fahad S, Ce dà A, Rizwan M, Fa had
W, Ehsan S, Bakha HF (2017) Ag o o es y: a sus ainable
en i onmen al p ac ice o ca bon seques a ion unde
he clima e change scena ios-a e iew. En i on Sci Pol-
lu Res In 24:11177–11191. h ps:// doi. o g/ 10. 1007/
s11356- 017- 8687-0
AdV (A bei sgemeinscha de Ve messungs e wal ungen de
Lände de Bundes epublik Deu schland) (2021) Doku-
men a ion zu Modellie ung de Geoin o ma ionen des
am lichen Ve messungswesens (GeoIn oDok) - E läu e -
ungen zum ATKIS® Basis-DLM - Ve sion 7.1.1. h ps://
www. ad - online. de/ icc/ ex deu/ na / 35b/ bina yw i e se
le ? imgUid= 3a130 147- 1420- 4e71- 0832- 12914 a39d 1 &
uBasV a ian = 11111 111- 1111- 1111- 1111- 11111 11111 11.
Accessed 12 Feb ua y 2025
AdV (A bei sgemeinscha de Ve messungs e wal ungen de
Lände de Bundes epublik Deu schland) (2024a) Ae ial
Pho og aphs. h ps:// www. ad - online. de/ P odu c s/ Geo o
pog a phy/ Ae ial- Pho o g aphs/. Accessed 5 June 2024
AdV (A bei sgemeinscha de Ve messungs e wal ungen
de Lände de Bundes epublik Deu schland) (2024b)
ATKIS® Digi al Landscape Models. h ps:// www. ad -
online. de/ P odu c s/ Geo o pog a phy/ Digi al- Lands cape-
Models/ Basis- DLM/. Accessed 12 Feb ua y 2025
Axe MS, G ange ID, Conway JS (2017) Ca bon s o age in
hedge biomass - a case s udy o ac i ely managed hedges
in England. Ag ic Ecosys En i on 250:81–88. h ps:// doi.
o g/ 10. 1016/j. agee. 2017. 08. 008
Ba á y P, Gallé R, Riesch F, Fische C, Do mann CF, Mußho
O, Császá P, Fusa o S, Gaye C, Happe A-K, Ku ucz K,
Molná D, Rösch V, Wie zke A, Tscha n ke T (2017) The
o me I on Cu ain s ill d i es biodi e si y-p o i ade-
o s in Ge man ag icul u e. Na Ecol E ol 1:1279–1284.
h ps:// doi. o g/ 10. 1038/ s41559- 017- 0272-x
Beck HE, Zimme mann NE, McVica TR, Ve gopolan N, Be g
A, Wood EF (2018) P esen and u u e Köppen-Geige
clima e classi ica ion maps a 1-km esolu ion. Scien i ic
Da a. h ps:// doi. o g/ 10. 1038/ sda a. 2018. 214
Black K, Lanigan G, Wa d M, Ka anagh I, hUallacháin DÓ,
Sulli an LO (2023) Biomass ca bon s ocks and s ock
changes in managed hedge ows. Sci To al En i on
871:162073. h ps:// doi. o g/ 10. 1016/j. sci o en . 2023.
162073
Buck O, Haub C, Wodi sch S, Lindemann D, Kleinewil-
linghö e L, Hazeu G, Kosz a B, Kleeschul e S, A nold
S, Hölzl M (2015) Task 1.9 - Analysis o he LUCAS
nomencla u e and p oposal o adap a ion o he nomen-
cla u e in iew o i s use by he Cope nicus land moni-
o ing se ices. Technical Repo . h ps:// land. cope nicus.
eu/ use - co ne / echn ical- lib a y/ LUCAS_ Cope nicus_
Repo _ 22. pd
Bundesam ü Ka og aphie und Geodäsie (2021) Ve wal-
ungsgebie e - NUTS-Gebie e 1:5000000, S and 31.12.
(NUTS5000 31.12.). h ps:// gdz. bkg. bund. de/ index. php/
de au l / digi ale- geoda en/ e wa l ung sgebi e e/ nu s- gebie
e-1- 5- 000- 000- s and- 31- 12- nu s5 000- 31- 12. h ml
Bundesminis e ium ü E näh ung und Landwi scha
(2022) Da en und Fak en: Land-, Fo s - und E näh ung-
swi scha mi Fische ei und Wein- und Ga enbau.
h ps:// www. bmel. de/ Sha e dDocs/ Downl oads/ DE/ B osc
hue en/ da en- ak en- 2022. pd . Accessed 14 Feb ua y
2025
Ag o o es Sys (2025) 99:267 Page 19 o 21 267
Vol.: (0123456789)
Bü ne G, Kosz a B, Soukup T, Sousa A, Langanke T (2017)
CLC 2018 Technical Guidelines. h ps:// land. cope nicus.
eu/ en/ echn ical- lib a y/ clc- 2018- echn ical- guide lines
Cas le SE, Mille DC, Me en N, O donez PJ, Baylis K (2022)
E idence o he impac s o ag o o es y on ecosys em
se ices and human well-being in high-income coun ies:
a sys ema ic map. En i on E id 11:10. h ps:// doi. o g/ 10.
1186/ s13750- 022- 00260-4
Ch is in S, He e É, Lecom e N (2019) Applica ions o deep
lea ning in ecology. Me hods Ecol E ol 10:1632–1644.
h ps:// doi. o g/ 10. 1111/ 2041- 210X. 13256
CLMS (Cope nicus Land Moni o ing Se ice) (2018) Use
manual - Small Woody Fea u es 2018 and Small Woody
Fea u es Changes 2015–2018
den He de M, Mo eno G, Mosque a-Losada RM, Palma JH,
Sidi opoulou A, San iago F eijanes JJ, C ous-Du an J,
Paulo JA, Tomé M, Pan e a A, Papanas asis VP, Man za-
nas K, Pachana P, Papadopoulos A, Plieninge T, Bu gess
PJ (2017) Cu en ex en and s a i ica ion o ag o o es y
in he Eu opean Union. Ag ic Ecosys En i on 241:121–
132. h ps:// doi. o g/ 10. 1016/j. agee. 2017. 03. 005
DG R d (Eu opean Commission: Di ec o a e-Gene al o
Resea ch and Inno a ion) (2021) Eu opean G een Deal
– Resea ch & inno a ion call. Publica ions O ice o he
Eu opean Union. h ps:// doi. o g/ 10. 2777/ 33415
EEA (Eu opean En i onmen Agency) (2020) Cope nicus
Land Moni o ing Se ice – High Resolu ion Laye Small
Woody Fea u es - 2015 e e ence yea . P oduc Speci ica-
ions & Use Guidelines
EEA (Eu opean En i onmen Agency) (2022) CLC+ Back-
bone P oduc Speci ica ion And Use Manual (Issue 3.0).
h ps:// land. cope nicus. eu/ en/ echn ical- lib a y/ clc- backb
one- p odu c - use - manual. Accessed 12 Feb ua y 2025
EEA (Eu opean En i onmen Agency) (2024a)
CLC+Backbone 2021 ( as e 10 m), Eu ope, 3-yea ly,
Jun. 2024
EEA (Eu opean En i onmen Agency) (2024b) CLCplus Back-
bone 2018 ( ec o ), Eu ope, No . 2024
EEA (Eu opean En i onmen Agency) (2024c) Woody land-
scape ea u es on ag icul u al land in Eu ope. h ps://
www. eea. eu opa. eu/ en/ analy sis/ indic a o s/ woody- lands
cape- ea u es- on- ag ic ul u al- land
Egli S, Höpke M (2020) CNN-based ee species classi ica ion
using high esolu ion RGB image da a om au oma ed
UAV obse a ions. Remo e Sens (Basel) 12:3892. h ps://
doi. o g/ 10. 3390/ s122 33892
Es i (2024) Wo ld Image y. h ps:// www. a cgis. com/ home/
i em. h ml? id= 10d 2 279 9 684e4 a9 6a 7 08 ebac2 a9.
Accessed 14 Feb ua y 2025
FAO (Food and Ag icul u e O ganiza ion) (2001) Global Fo -
es Resou ces Assessmen 2000. h ps:// www. ao. o g/4/
y1997e/ y1997 e00. h m
Fede al Agency o Na u e Conse a ion (2025) Moni o ing o
high na u e alue a mland. h ps:// www. b n. de/ en/ moni
o ing- high- na u e- alue- a ml and# ancho - 2331. Accessed
12 Feb ua y 2025
Funk R, Völke L, Deumlich D (2023) Landscape s uc-
u e model based es ima ion o he wind e osion isk in
B andenbu g, Ge many. Aeolian Res 62:100878. h ps://
doi. o g/ 10. 1016/j. aeolia. 2023. 100878
Golicz K, Ghaza yan G, Nie he W, Wa enbe g AC, B eue
L, Ga inge A, Jacobs SR, Kleinebecke T, Weckenb ock
P, G oße-S ol enbe g A (2021) The ole o small woody
landscape ea u es and ag o o es y sys ems o na ional
ca bon budge ing in Ge many. Land (Basel) 10:1028.
h ps:// doi. o g/ 10. 3390/ land1 01010 28
Golicz K, Belling a h-Kimu a S, B eue L, Wa enbe g AC
(2022) Ca bon accoun ing in Eu opean ag o o es y sys-
ems – key esea ch gaps and da a needs. Cu Res En i-
on Sus ain 4:100134. h ps:// doi. o g/ 10. 1016/j. c sus .
2022. 100134
Google (2023) Map Da a. h ps:// www. google. com/ maps
G een S, Ma in S, Gha echelou S, Cawkwell F, Black K
(2021) B ia : biomass e ie al in I eland using ac i e
emo e sensing (2014-Cc p-Ms.17). h ps:// www. epa. ie/
publi ca io ns/ esea ch/ clima e- change/ Resea ch_ Repo _
305. pd . Accessed 14 Feb ua y 2025
Gues G, B igh RM, Che ubini F, S ømman AH (2013) Con-
sis en quan i ica ion o clima e impac s due o biogenic
ca bon s o age ac oss a ange o bio-p oduc sys ems.
En i on Impac Assess Re 43:21–30. h ps:// doi. o g/ 10.
1016/j. eia . 2013. 05. 002
He nández-Mo cillo M, Bu gess P, Mi ck J, Pan e a A,
Plieninge T (2018) Scanning ag o o es y-based solu-
ions o clima e change mi iga ion and adap a ion in
Eu ope. En i on Sci Policy 80:44–52. h ps:// doi. o g/ 10.
1016/j. en sci. 2017. 11. 013
He zog F (1998) S euobs : a adi ional ag o o es y sys-
em as a model o ag o o es y de elopmen in empe -
a e Eu ope. Ag o o Sys 42:61–80. h ps:// doi. o g/ 10.
1023/A: 10061 52127 824
Jäge E (2003) ATKIS® als Gemeinscha sau gabe de Lände
und des Bundes. j. Ca og Geog In 53:113–119. h ps://
doi. o g/ 10. 1007/ BF035 45526
Jose S (2009) Ag o o es y o ecosys em se ices and en i on-
men al bene i s: an o e iew. Ag o o es Sys 76(1):1–10.
h ps:// doi. o g/ 10. 1007/ s10457- 009- 9229-7
Jo anelly K, Su C, Ong TW (2025) Managing ag o o es y
ansi ions in a apidly changing clima e. Ag oecol Sus-
ainable Food Sys 49:228–268. h ps:// doi. o g/ 10. 1080/
21683 565. 2024. 24160 09
Kompe enzzen um Ökolandbau Niede sachsen (2019)
S euobs wiesen p lanzen: A en eiche Lebens äume und
„Regenwälde Deu schlands“. h ps:// o en bu g. bund. ne /
ilea dmin/ o en bu g/ Landn a u / koen_ s eu obs w iese_
p lan zen. pd . Accessed 19 Feb ua y 2025
Kuma S, Alam B, Ta ia S, Singh P, Yada A, A unachalam
A (2024) Ag o o es y solu ions o clima e change and
en i onmen al es o a ion. Sp inge Na u e Singapo e,
Singapo e. h ps:// doi. o g/ 10. 1007/ 978- 981- 97- 5004-7
Lal Sh es ha H (2020) Assessmen o abo e g ound biomass
o T ees ou side Fo es (TOF) in he Con ex o Clima e
Change. JENR 4. h ps:// doi. o g/ 10. 23880/ JENR- 16000
186
Lechne AM, S ein A, Jones SD, Fe we da JG (2009) Remo e
sensing o small and linea ea u es: quan i ying he
e ec s o pa ch size and leng h, g id posi ion and de ec -
abili y on land co e mapping. Remo e Sens En i on
113:2194–2204. h ps:// doi. o g/ 10. 1016/j. se. 2009. 06. 002
Lingne S, Thiessen E, Ha ung E (2018) Abo eg ound bio-
mass es ima ion in linea o es objec s: 2D- s. 3D-da a.
Ag o o es Sys (2025) 99:267
267 Page 20 o 21
Vol:. (1234567890)
J Fo es Sci 64:523–532. h ps:// doi. o g/ 10. 17221/ 106/
2018- JFS
Liu S, B and M, No d-La sen T, Cha e J, Reine F, Lang N,
Tong X, Ciais P, Igel C, Pascual A, Gue a-He nandez
J, Li S, Mugabowindekwe M, Saa chi S, Yue Y, Chen Z,
Fenshol R (2023) The o e looked con ibu ion o ees
ou side o es s o ee co e and woody biomass ac oss
Eu ope. Sci Ad 9:eadh4097. h ps:// doi. o g/ 10. 1126/ sci-
ad . adh40 97
Nai PKR (2012) Ca bon seques a ion s udies in ag o o -
es y sys ems: a eali y-check. Ag o o Sys 86:243–253.
h ps:// doi. o g/ 10. 1007/ s10457- 011- 9434-z
Nai PKR, Kuma BM, Nai VD (2021) De ini ion and Con-
cep s o Ag o o es y. In: Nai PKR, Kuma BM, Nai VD
(eds) An In oduc ion o Ag o o es y. Sp inge In e na-
ional Publishing, Cham, pp 21–28. h ps:// doi. o g/ 10.
1007/ 978-3- 030- 75358-0_2
Pan e a Α, Mosque a-Losada MR, He zog F, den He de
M (2021) Ag o o es y and he en i onmen . Ag o-
o es Sys 95(5):767–774. h ps:// doi. o g/ 10. 1007/
s10457- 021- 00640-8
Pa don P, Me ens J, Reubens B, Reheul D, Coussemen T, Elsen
A, Nelissen V, Ve heyen K (2020) Juglans egia (walnu ) in
empe a e a able ag o o es y sys ems: e ec s on soil cha -
ac e is ics, a h opod di e si y and c op yield. Renewable
Ag ic Food Sys 35:533–549. h ps:// doi. o g/ 10. 1017/ S1742
17051 90001 76
Pa ikh D, Zi nick CL (2010) The ole o ea u es, algo i hms
and da a in isual ecogni ion. 2010 IEEE Compu e
Socie y Con e ence on Compu e Vision and Pa e n
Recogni ion:2328–2335
Plieninge T (2011) Capi alizing on he ca bon seques a ion
po en ial o ag o o es y in Ge many’s ag icul u al land-
scapes: ealigning he clima e change mi iga ion and land-
scape conse a ion agendas. Landsc Res 36:435–454.
h ps:// doi. o g/ 10. 1080/ 01426 397. 2011. 582943
QGIS De elopmen Team (2022) QGIS Geog aphic In o ma ion
Sys em. h ps:// www. qgis. o g/
R Co e Team (2025) R: A Language and En i onmen o S a is-
ical Compu ing. h ps:// www.R- p oje c . o g/
Reisne Y, de Filippi R, He zog F, Palma J (2007) Ta ge egions
o sil oa able ag o o es y in Eu ope. Ecol Eng 29:401–
418. h ps:// doi. o g/ 10. 1016/j. ecole ng. 2006. 09. 020
Rubio-Delgado J, Schnabel S, La ado-Con ado JF, Schmu z U
(2024) Small woody ea u es in ag icul u al a eas: ag o o -
es y sys ems o o e looked signi icance in Eu ope. Ag ic
Sys 218:103973. h ps:// doi. o g/ 10. 1016/j. agsy. 2024.
103973
Sa le C, Sch ade J, Hü ne M-L, Henle K (2024) E ec s o
managemen , habi a and landscape cha ac e is ics on bio-
di e si y o o cha d meadows in Cen al Eu ope: a b ie
e iew. Na Conse 55:103–134. h ps:// doi. o g/ 10. 3897/
na u econs e a ion. 55. 108688
Scha e C, Elbakidze M, Bjö klund J (2024) Mo i a ion and pe -
cep ion o a me s on he bene i s and challenges o ag o o -
es y in Sweden (No he n Eu ope). Ag o o Sys 98:939–
958. h ps:// doi. o g/ 10. 1007/ s10457- 024- 00964-1
Schie e F, Ka enbo n T, F ick A, F ey J, Schall P, Koch B,
Schmid lein S (2020) Mapping o es ee species in high
esolu ion UAV-based RGB-image y by means o con olu-
ional neu al ne wo ks. ISPRS J Pho og amm Remo e Sens
170:205–215. h ps:// doi. o g/ 10. 1016/j. isp s jp s. 2020. 10.
015
Schindle Z, Sei e T, Sheppa d JP, Mo ha C (2023a) Allome -
ic models o abo e-g ound biomass, ca bon and nu ien
con en o wild che y (P unus a ium L.) ees in ag o o -
es y sys ems. Ann o Sci 80:28. h ps:// doi. o g/ 10. 1186/
s13595- 023- 01196-6
Schindle Z, Mo ha C, Sheppa d JP, F ey J, Sei e T (2023b)
In a nu shell: explo ing single ee pa ame e s and abo e-
g ound ca bon seques a ion po en ial o common walnu
(Juglans egia L.) in ag o o es y sys ems. Ag o o Sys
97:1007–1024. h ps:// doi. o g/ 10. 1007/ s10457- 023- 00844-0
Schnell S, Kleinn C, S åhl G (2015) Moni o ing ees ou side o -
es s: a e iew. En i on Moni Assess 187:600. h ps:// doi.
o g/ 10. 1007/ s10661- 015- 4817-7
Schumache U, Meinel G (2009) ATKIS, ALK(IS), O hobild –
Ve gleich on Da eng undlagen eines Flächenmoni o ings.
In: Schumache U, Meinel G (eds) Flächennu zungsmoni-
o ing. 1: Konzep e, Indika o en, S a is ik. Shake , pp 47–67
Seidel D, Busch G, K ause B, Bade C, Fessel C, Kleinn C
(2015) Quan i ica ion o biomass p oduc ion po en ials
om ees ou side o es s - a case s udy om Cen al Ge -
many. Bioene gy Res 8:1344–1351. h ps:// doi. o g/ 10. 1007/
s12155- 015- 9596-z
Sha ma P, Bha dwaj DR, Singh MK, Nigam R, Pala NA, Kuma
A, Ve ma K, Kuma D, Thaku P (2023) Geospa ial ech-
nology in ag o o es y: s a us, p ospec s, and cons ain s.
En i on Sci Pollu Res In 30:116459–116487. h ps:// doi.
o g/ 10. 1007/ s11356- 022- 20305-y
Sheppa d JP, La ysch E, Cua anhua CJ, Schindle Z, Du Toi B,
Malhe be GF, Kunneke A, Mo ha C, Bohn Reckziegel R,
Sei e T, Kahle H-P (2024) Assessmen o biomass and
ca bon s o age o a Populus simonii windb eak loca ed in
he Wes e n Cape P o ince, Sou h A ica. Ag o o es Sys
98:697–714. h ps:// doi. o g/ 10. 1007/ s10457- 023- 00940-1
Sollen-No lin M, Ghaley BB, Rin oul NLJ (2020) Ag o o es y
bene i s and challenges o adop ion in Eu ope and beyond.
Sus ainabili y 12:7001. h ps:// doi. o g/ 10. 3390/ su121 77001
S a is isches Bundesam G-O (2024) Boden läche ( a sächli-
che Nu zung). h ps:// www- genes is. des a is. de/ da en bank/
online/ s a i s ic/ 33111/ able/ 33111- 0005. Accessed 22 Ap il
2024
Te asaki Ha DE, Yeo S, Alma az M, Beillouin D, Ca dinael R,
Ga cia E, Kay S, Lo ell ST, Rosens ock TS, Sp enkle-Hyp-
poli e S, S olle F, Sube M, Thapa B, Wood S, Cook-Pa on
SC (2023) P io i y science can accele a e ag o o es y as a
na u al clima e solu ion. Na Clim Chang 13:1179–1190.
h ps:// doi. o g/ 10. 1038/ s41558- 023- 01810-5
Thapa B, Lo ell S, Wilson J (2023) Remo e sensing and machine
lea ning applica ions o abo eg ound biomass es ima ion in
ag o o es y sys ems: a e iew. Ag o o Sys 97:1097–1111.
h ps:// doi. o g/ 10. 1007/ s10457- 023- 00850-2
Thomas SC, Ma in AR (2012) Ca bon con en o ee issues: a
syn hesis. Fo es s 3:332–352. h ps:// doi. o g/ 10. 3390/ 3020
332
Udawa a RP, Ranko h L, Jose S (2019) Ag o o es y and biodi-
e si y. Sus ainabili y 11(10):2879. h ps:// doi. o g/ 10. 3390/
su111 02879
Wo ld Bank (2011) Clima e Change Knowledge Po al: Obse ed
Clima e Da a, CRU Ts4.07 0.5-Deg ee
Ag o o es Sys (2025) 99:267 Page 21 o 21 267
Vol.: (0123456789)
Zome RJ, Neu eld H, Xu J, Ah ends A, Bossio D, T abucco A,
an Noo dwijk M, Wang M (2016) Global ee co e and
biomass ca bon on ag icul u al land: he con ibu ion o
ag o o es y o global and na ional ca bon budge s. Sci Rep
6:29987. h ps:// doi. o g/ 10. 1038/ s ep2 9987
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d
o ju isdic ional claims in published maps and ins i u ional
a ilia ions.