Connec ing habi a s in Eu opean ag icul u al landscapes: Fa me s’spa ial
p e e ences o linea wildli e co ido s
Fabian Klebl
a,b,*
, Jona han R. Rhodes
c
, Ka i H¨
a ne
a
, Anne e Pio
a
a
Leibniz Cen e o Ag icul u al Landscape Resea ch (ZALF), Ebe swalde S aße 84, Münchebe g 15374, Ge many
b
Ag icul u al and Food Policy G oup, Alb ech Daniel Thae Ins i u e o Ag icul u al and Ho icul u al Sciences, Humbold -Uni e si ¨
a zu Be lin, Un e den Linden 6,
Be lin 10099, Ge many
c
School o Biology and En i onmen al Science, Queensland Uni e si y o Technology, B isbane, Queensland 4000, Aus alia
HIGHLIGHTS
•Fa me s p e e o place measu es o minimise dis u bance and maximise p oduc i i y.
•Landscape cha ac e is ics guide a me s’decisions.
•Habi a connec i i y is a ele an concep o a me s.
•Local a ming adi ions and p ac ices a e c ucial.
•The s udy esul s can help o educe ansac ion cos s by a ge ing likely adop e s.
ARTICLE INFO
Keywo ds:
Biodi e si y conse a ion
Habi a agmen a ion
Landscape connec i i y
Ecological co ido s
Hedge ows
Ag i-en i onmen al measu es
ABSTRACT
Habi a agmen a ion in ag icul u al landscapes h ea ens biodi e si y. Enhancing landscape connec i i y ac oss
cul i a ed a eas equi es a ho ough unde s anding o a me s’spa ial conside a ions and hei willingness o
c ea e semi-na u al habi a s. We he e o e conduc ed a spa ial choice expe imen wi h a me s om en Eu o-
pean coun ies o assess hei p e e ences o placing linea wildli e habi a s (hedge ows and wild lowe s ips) a
he ield scale unde di e en scena ios, as well as he ole o a m and pe sonal ac o s. A o al o 471 esponses
we e analysed using mul inomial logis ic eg ession and gene alised linea mixed models. The esul s indica e
ha landscape condi ions, including ield shape, slope, soil quali y, and p e-exis ing landscape ea u es, exe a
signi ican in luence on a me s’decisions, as do he size o machine y, cul u al egions, a i udes owa ds
biodi e si y, and ype o in e en ion. On he o he hand, no s a is ical signi icance was ound o o he a i-
ables. In gene al, a me s’choices we e d i en by a desi e o minimise dis u bance o ield wo k, op imise
p oduc i i y, inc ease biodi e si y, and add ess speci ic en i onmen al challenges. The insigh s in o a me s’
decision-making om his s udy can in o m ecological ne wo k planning o educe ansac ion cos s by p e-
selec ing likely adop e s, and o mi iga e esis ance and lowe inancial compensa ion by iden i ying bes - i
op ions aligned wi h a me s’p ac ices. In eg a ing hese indings in o geospa ial models could imp o e p e-
dic ions o he impac o spa ially a ge ed biodi e si y conse a ion s a egies on landscape composi ion and
u u e biodi e si y ends in ag icul u al a eas.
1. In oduc ion
Habi a agmen a ion poses an u gen h ea o biodi e si y in
ag icul u al a eas and is a ec ed by how a me s manage hei land (e.
g., Haddad e al., 2015; Schlaep e e al., 2018). The ( e)in oduc ion o
wildli e co ido s on a mland o connec exis ing habi a s is seen as an
impo an means o mi iga e nega i e e ec s o habi a and species
isola ion. These co ido s can ake adi ional o ms, such as hedge ows,
o mo e con empo a y app oaches, such as pe ennial wild lowe s ips.
Ye , he e is s ill limi ed unde s anding o a me s’pe spec i es on
landscape connec i i y and hei decisions o con ibu e o i .
Hedge ows in pa icula ha e demons a ed conside able po en ial
o enhancing habi a connec i i y. While he speci ic e ec s a y
depending on he landscape and species unde s udy, he e is e idence
ha hedge ows in ag icul u al landscapes ac as co ido s o nume ous
species, enabling hei mo emen and gene low be ween small, isola ed
* Co esponding au ho a : Ebe swalde S aße 84, Münchebe g 15374, Ge many.
E-mail add ess: [email p o ec ed] (F. Klebl).
Con en s lis s a ailable a ScienceDi ec
Landscape and U ban Planning
jou nal homepage: www.else ie .com/loca e/landu bplan
h ps://doi.o g/10.1016/j.landu bplan.2025.105325
Recei ed 12 Augus 2024; Recei ed in e ised o m 17 Decembe 2024; Accep ed 13 Feb ua y 2025
Landscape and U ban Planning 258 (2025) 105325
A ailable online 1 Ma ch 2025
0169-2046/© 2025 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (
h p://c ea i ecommons.o g/licenses/by/4.0/ ).
habi a pa ches (Dondina e al., 2016; Fische e al., 2013; Gelling e al.,
2007; K a schme e al., 2024; Licca i e al., 2022; Michel e al., 2006;
Moo house e al., 2014; Pelle ie -Gui ie e al., 2020; Po ela e al.,
2020; T a e s e al., 2021; Vasilie &G eenwood, 2023; Wehling &
Diekmann, 2009). Despi e he isk ha ecological co ido s may acili-
a e he sp ead o diseases and in asi e species (Mon gome y e al.,
2020), hey emain a i al elemen in landscapes wi h small habi a
pa ches whe e popula ions a e uns able and habi a expansion is no
easible (Donaldson e al., 2017).
Howe e , es ablishing hedge ows equi es subs an ial e o and a
long- e m commi men om a me s, o en spanning mul iple decades.
As a ade-o , his s udy also conside s pe ennial wild lowe s ips o
he c ea ion o linea connec ing habi a s. Al hough hei o e all impac
on landscape connec i i y is no as p onounced as ha o hedge ows,
hey o e con inuous esou ces and e uges o a ious species, hus
suppo ing hei dispe sal and su i al in o he wise inhospi able ag i-
cul u al a eas (A i on e al., 2011; Szi ´
a e al., 2022). The s a egic
spa ial placemen o hese s ips by a me s is c ucial o de eloping a
comp ehensi e lowe ing ne wo k ac oss en i e landscapes, he eby
maximising hei ecological bene i s (Buhk e al., 2018).
To implemen a ne wo k o connec ed habi a s in ag icul u al a eas,
conse a ion s a egies need o be based on an unde s anding o wha
mo i a es a me s o c ea e hese habi a s and o connec hem o he
ne wo k. While much is known abou a me s’gene al mo i a ions o
adop ing biodi e si y- iendly p ac ices (Klebl, Feind , &Pio , 2024a),
o he bes o ou knowledge, no s udy has ye assessed whe e a me s
a e mos likely o place biodi e si y measu es a he ield scale, wha
in luences hei unde lying decisions, o whe he habi a connec i i y is
a ele an conside a ion o hem.
This s udy in es iga es he ac o s in luencing a me s’decisions
abou he placemen o biodi e si y measu es. Speci ically, i examines
wha d i es a me s’choices in he spa ial alloca ion o linea habi a s
such as hedge ows and lowe s ips, and how ield and landscape
cha ac e is ics impac hei p e e ences. These ques ions we e add essed
h ough a spa ial choice expe imen in which a me s esponded o
hypo he ical scena ios. The esul ing da a we e s a is ically analysed o
e alua e he in luence o ac o s ela ed o he a me s and hei a ms,
as well as he hypo he ical ield and landscape se ings.
2. Ma e ial and me hods
2.1. S udy a eas and sample
The p esen s udy elies on a su ey conduc ed among a me s ac oss
se e al Eu opean coun ies (Fig. 1), ep esen ing a di e se ange o
landscapes, clima e condi ions, a ming sys ems, and socio-cul u al
con ex s. These coun ies we e selec ed o hei ole as hos s o he
ecological and socio-economic expe imen al si es wi hin he SHOW-
CASE esea ch p ojec (h ps://showcase-p ojec .eu/). To ob ain a
ep esen a i e sample, a me s we e andomly ec ui ed h ough mul-
iple channels, including a me s’associa ions, NGOs, and local
ne wo ks.
2.2. Da a collec ion
Fa me s we e i s asked o p o ide gene al pe sonal in o ma ion,
including age and educa ional backg ound, along wi h de ailed a m
cha ac e is ics, such as a m size, ope a ional ocus, a e age ield size,
ield ope a ions, machine y size, and landscape condi ions. Like scales
we e employed o e alua e a me s’a i udes owa ds biodi e si y,
assessing he impo ance hey a ibu e o i and i s conse a ion, bo h in
gene al and in ag icul u al landscapes.
The su ey’s co e componen was a spa ial choice expe imen aimed
a cap u ing a me s’p e e ences o alloca ing biodi e si y measu es a
he ield scale. Pa icipan s we e andomly assigned o one o wo
g oups: one ocused on plan ing a wild lowe s ip as a empo a y
in e en ion (minimum wid h o ou me es, las ing a leas h ee
yea s), while he o he conside ed es ablishing a na u ally g own
hedge ow as a long- e m landscape elemen (minimum heigh and wid h
o wo me es each).
Responden s we e hen in i ed o en ision a hypo he ical ield (o
meadow o pas u e, depending on hei ope a ional ocus) ha is
accessible om all sides. In he baseline scena io, he ield was desc ibed
as being o a e age size as epo ed by he a me , wi h a ec angula
shape, la su ace, and uni o m soil quali y. In each subsequen sce-
na io, one cha ac e is ic was al e ed while he o he s emained consis-
en wi h he baseline. These al e a ions included he in oduc ion o a
slope, a soil quali y g adien , a di e en ield shape, doubling he ield
size, a p e ailing wind di ec ion, as well as a ying se ings such as
p oximi y o a o es o oad, and he p esence o exis ing hedges on
neighbou ing ields (Table 1).
In each scena io, pa icipan s selec ed hei i s and second p e e -
ences om he a ailable op ions by clicking on co esponding g aphical
illus a ions and b ie ly explaining hei choice. Fo example, in he
baseline scena io, pa icipan s could choose om ou op ions, as shown
in Fig. 2. Al hough second p e e ences we e eco ded o gain addi ional
insigh s, hey we e excluded om he inal analysis as no sound
app oach was iden i ied o weigh o inco po a e he second choice in a
meaning ul way. Fu he de ails on he su ey can be ound in he
supplemen a y ma e ial.
The su ey was pa o a b oade p ojec su ey and was designed as
an online applica ion, accessible ia web b owse s and mobile de ices.
Howe e , in some egions o Romania, whe e he online o ma was
unsui able, a p in ed e sion was used and comple ed du ing wo kshops
o ganised by local p ojec pa ne s. Da a collec ion occu ed be ween
Janua y and Oc obe 2023, yielding 471 alid esponses.
2.3. Da a analysis
2.3.1. Da a p epa a ion
The da ase con ains ela i ely ew missing alues, as he choices and
mos su ey ques ions we e manda o y. Missing da a, classi ied as
Fig. 1. O e iew o s udy a eas: Es onia, F ance, Hunga y, Po ugal, Romania,
Spain, Sweden, Swi ze land, he Ne he lands, and he UK.
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
2
‘missing comple ely a andom’, we e add essed h ough he Mul i a -
ia e Impu a ion by Chained Equa ions (MICE) me hod. Linea p edic o s
we e impu ed ia p edic i e mean ma ching, whe eas ca ego ical alues
we e es ima ed using logis ic eg ession. An i e a i e app oach was
employed o achie e a inal impu a ion ha closely ma ched he o iginal
da ase , u ilising he {mice} package ( an Buu en &G oo huis-
Oudshoo n, 2011) in R e sion 4.4.1 (R Co e Team, 2024). The da a
densi y be o e and a e impu a ion exhibi ed an almos iden ical
pa e n (Fig. A1 in he Appendix A), indica ing a high accu acy o he
impu ed da a.
Pa icipan s’ ex ual explana ions o hei choices we e ansla ed
and ca ego ised. They we e no analysed quan i a i ely due o equen
omissions o i ele ance. Ne e heless, an o e iew o he mos
commonly ci ed easons is p o ided, which o e s quali a i e insigh s
ha help o con ex ualise he decision-making p ocess.
2.3.2. Mul inomial logis ic eg ession model
Since he choice op ions a e ca ego ical ou comes wi h mul iple
possible ca ego ies, we employed a mul inomial logis ic eg ession
model i ed o each scena io o in es iga e he ela ionship be ween
se e al p edic o a iables and he esponse a iable (Ag es i, 2007).
The model can be desc ibed as ollows:
log(
π
j
π
J)=β0j+β1jX1+β2jX2⋯+βkjXk(1)
o j=1,⋯,J−1, whe e
π
jis he p obabili y o he ou come (i.e., he
choice) being in ca ego y jand
π
Jin he e e ence ca ego y (J), β0jis he
in e cep o ca ego y j, and βij a e he coe icien s o he p edic o
a iables Xi.
As log-odds a e unin ui i e o in e p e , we de i ed he odds a io
(OR) o es ima e he change in he odds o being in a speci ic ca ego y o
he ou come a iable wi h a one-uni inc ease in he p edic o a iable,
ela i e o he e e ence ca ego y (Hosme e al., 2013). This is de ined
as
θij =e
ˆ
βij (2)
In o mula ing he eg ession models, se e al a iables we e ound no
o signi ican ly enhance he model’s explana o y powe bu o inc ease
i s complexi y. These a iables included a m size, he p opo ion o
owned land o o al cul i a ed land, he ope a ional s a us o he a m
( ull- ime o pa - ime), p io implemen a ion o simila in e en ions (i.
e., hedge ows o wild lowe s ips), a me s’gene al and ag icul u al
educa ion le els, and a me s’age. Consequen ly, hese ac o s we e
excluded om he inal models. The a iables ha p o ided an
explana o y con ibu ion and/o we e an impo an componen o he
ini ial assump ions we e e ained as independen a iables and a e
summa ised in Table 2.
Due o a high co ela ion be ween he coun ies and p edic o a i-
ables, i was imp ac ical o include coun y as an independen a iable.
Howe e , egional in luences we e conside ed ele an beyond hese
ac o s. The e o e, a socio-cul u al ca ego isa ion was conduc ed,
emphasising majo ag icul u al signi icance. The egions we e de ined
as Medi e anean a ming sys ems in sou he n Eu ope (ES, PT), Eu o-
pean coun ies o me ly aligned wi h he So ie Union (EE, HU, RO),
and wes e n and no he n Eu opean coun ies (CH, FR, NL, SE, UK),
which se ed as he e e ence g oup.
The eg ession models a e ounded on ce ain key assump ions. All
obse a ions in he sample a e deemed o be independen and su icien
in numbe , wi h a leas abou en obse a ions pe p edic o a iable o
each ou come ca ego y. Ca ego ies ha did no mee hese condi ions
we e excluded.
1
The independence o i ele an al e na i es (IIA) hy-
po hesis was es ed using he Hausman-McFadden Tes ia he mlogi ::
hm es () unc ion and was ejec ed o each a iable. The Box-Tidwell
es (p=0.46) con i med he assump ion o linea i y be ween
Table 1
O e iew o he scena ios in he spa ial choice expe imen .
Scena io Sample
sc00 baseline F/H
sc01 slope (ho izon al g adien ) F/H
sc02 soil quali y (ho izon al g adien ) F/H
sc03 soil quali y ( e ical g adien ) F/H
sc04 shape F
sc05 inc ease in size F
sc06 wind H
sc07 o es F/H
sc08 o es F/H
sc09 oad H
sc10 exis ing hedges H
sc11 exis ing hedges H
sc12 exis ing hedges H
The columns p esen he scena io code and name, along wi h a g aphical illus-
a ion, and indica e he sample (wild lowe s ip (F) o hedge ow (H)) o which
he scena io applies.
Fig. 2. Choice op ions in he baseline scena io. The black ba indica es he
biodi e si y measu e in he ield.
1
i.e., [scena io code −choice op ion] sc00-D, sc04-D, sc04-B1, sc06-C/D,
sc11-B2.
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
3
con inuous p edic o s and he log-odds o ou come ca ego ies.
2
Wi h a Cox’s dis ance alue o 359.4 a 393 deg ees o eedom, he
model’s i was decided o be sa is ac o y, sugges ing a easonable
alignmen be ween p edic ed p obabili ies and obse ed da a. Mul i-
collinea i y was assessed using a a iance in la ion ac o (VIF) ole -
ance o <5, and le els we e gene ally e y low.
3
Finally, he
app oxima ely symme ic dis ibu ion o esiduals in each model
con i med he eliabili y and sui abili y o he model o analysis.
The calcula ion o he log-odds o each choice op ion ela i e o he
p edic o a iables was p ima ily based on he {nne } package, com-
bined wi h addi ional es s equi ing he {mlogi } package, such as he
Hausman-McFadden es . To de e mine he signi icance o indi idual
eg ession coe icien s, a Wald es was pe o med using he {lm es }
package, h ough which he zs a is ics we e ob ained o each pa am-
e e . The o e all ele ance o each a iable was assessed using a
likelihood- a io es , which is pa icula ly eliable o small sample sizes
(Ag es i, 2007).
2.3.3. Gene alised linea mixed model (GLMM)
In o de o e alua e he impac o di e en scena ios on he de-
cisions, choices om each scena io we e combined wi h hose om he
baseline, and a dummy a iable (SCENARIO) was in oduced o indica e
whe he a choice occu ed unde a speci ic scena io o he baseline
condi ions. As each esponden p o ided da a o bo h he scena io and
he baseline, which iola es he assump ion o independen obse a-
ions, a GLMM was employed. This app oach accoun ed o epea ed
measu es by inco po a ing andom in e cep s o each esponden ,
he eby con olling o indi idual a iabili y in he analysis (Ag es i,
2007).
The logi link unc ion o he log-odds o he selec ed choice op ion
in a gi en scena io ela i e o he baseline is:
log(P(Yij =1)
1−P(Yij =1))=β0+β1Xij +
μ
0i(3)
whe e Yij is he bina y esponse a iable o he indi idual iand
obse a ion j,β0is he ixed in e cep , β1is he ixed e ec coe icien o
he p edic o Xij (i.e., SCENARIO), and
μ
0iis he andom e ec o i,
which ollows a no mal dis ibu ion
μ
0iN(0,
σ
2
μ
).
The models we e de eloped using he lme4::glme () unc ion o
es ing each choice op ion (n =37) agains he a iable SCENARIO as a
ixed e ec , wi h a andom e ec added o esponden ID. To ensu e
consis ency wi h he symme ical baseline condi ions, sub-choices
ele an only o speci ic scena ios we e agg ega ed and s anda dised
(e.g., me ging he uppe (B1) and lowe (B2) edge op ions in he slope
scena io in o B). Odds a ios we e calcula ed as desc ibed abo e.
2.4. Limi a ions and po en ial bias
A challenge inhe en o his s udy was he ela i ely limi ed sample
size compa ed o he numbe o choice op ions, which led o some op-
ions being a ely selec ed and subsequen ly excluded om analysis.
Mo eo e , he e is a possibili y o misin e p e a ion in he choice
expe imen , since some a me s assumed ac o s such as a no h–sou h
alignmen o egion-speci ic wind di ec ion. When hese assump ions
we e iden i ied h ough w i en esponses, he obse a ions we e
emo ed om he analysis. Howe e , i is likely ha some esponden s
made implici assump ions wi hou s a ing hem, which could ha e
in luenced he esul s.
3. Resul s
3.1. Da a o e iew
O he 471 a me s who comple ed he su ey, 55 % we e p ima ily
engaged in a able a ming, 17 % in meadow managemen , and 28 % in
g azing sys ems. Fa m sizes anged widely om 0.03 ha o 9200 ha, wi h
a median o 70 ha. The majo i y o esponden s (64 %) we e ull- ime
a me s who owned mos o he land hey wo ked on. The p opo ion
o o ganic a ms in his sample was disp opo iona ely high a 42 %. A
de ailed o e iew o he sample and p edic o a iables is a ailable in
he supplemen a y ma e ial.
Decisions on he placemen o biodi e si y measu es showed dis inc
pa e ns. Unde he baseline condi ions (sc00), op ion A (leng hwise a
he edge) was mos equen ly chosen due o minimal ope a ional
dis u bance (Fig. 3). This is consis en wi h placing amlines leng hwise
o a oid obs uc ing machine y u ning a he sho edges. Op ion B
(wid hwise a he edge) was also popula , as i minimised bo h dis u -
bance and a ea used o he measu e. Op ion C (leng hwise in- ield) was
selec ed less o en, mainly o i s po en ial o p omo e biodi e si y while
allowing o leng hwise managemen . Op ion D (wid hwise in- ield) was
a ely chosen, ypically o di iding he ield in o sepa a e plo s o
pas u e pa cels.
The choices ac oss all scena ios a e summa ised in Fig. A2 in he
Appendix A. In he slope scena io (sc01), op ions B1 (wid hwise a he
Table 2
Desc ip ion o he p edic o a iables included in he mul inomial logis ic
eg ession models.
Va iable Desc ip ion Values
ATT_BIODIV a i udes owa ds biodi e si y Like scale (1–10)
FIELDSIZE a e age ield size [num]
LU p edominan land use o
a mland
a able land/g assland
ORGANIC o ganic ce i ica ion binomial
REGION egions o cul u al, his o ical,
and geophysical simila i ies
ele an o a ming
Wes and No h Eu ope (1),
o me USSR-aligned coun ies
(2), Medi e anean (3)
SAMPLE sample alloca ion lowe s ip (F)/hedge ow (H)
RELIEF opog aphic elie la /hilly
WORKWIDTH maximum wo k wid h o
machine y
[num]
Fig. 3. Rela i e dis ibu ion o a me s’choices o he spa ial alloca ion o
hedge ows (H) and wild lowe s ips (F) in he baseline scena io. The y-axis
labels deno e he di e en choice op ions o each scena io, wi h g aphics
ep esen ing he biodi e si y measu e as a black ba in he ield. The plo labels
show he mos equen ly ci ed easons o each choice.
2
The e e ence alues a e based on he baseline scena io bu we e also
e alua ed unde each scena io.
3
The only excep ions a e a mode a e collinea i y o FIELDSIZE in sc01 (VIF
=6.3), and o WORKWIDTH (VIF =5.6) and REGION (VIF =6.9) in sc02.
Howe e , his did no subs an ially a ec model pe o mance.
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
4
highe edge) and B2 (wid hwise a he lowe edge) we e mos equen ly
selec ed due o e osion con ol conce ns. Howe e , he a ionales
di e ed: B1 was aimed a p e en ing un-o wa e om en e ing he
ield, while B2 was chosen o e ain wa e and nu ien s wi hin he ield.
In he wind scena io (sc06), a me s p io i ised placing hedge ows o
mi iga e wind e ec s, hough some a oided wind ba ie s o ensu e
p ope d ying o hay o lea es, o o acili a e wind pollina ion. In he
soil quali y scena ios (sc02 and sc03), measu es we e gene ally placed in
a eas wi h he lowes soil quali y. A simila pa e n was obse ed in he
o es scena ios (sc07 and sc08), whe e p oximi y o he o es was
chosen in conside a ion o educed p oduc i i y om shading.
When in eg a ing he hypo he ical ield in o landscape cons ella ions,
many a me s sough o u ilise biodi e si y habi a s as a means o p e-
en ing con amina ion om o he ields (sc09-sc12), pollu ion and noise
om oads (sc09), o wildli e en e ing om he o es (sc07 and sc08).
While many a me s in ended o inc ease biodi e si y in he o es sce-
na ios, pe cep ions o he bes choice o inc easing biodi e si y a ied.
Some p e e ed op ions close o he o es o habi a connec i i y,
whe eas o he s a ou ed op ions a he om he o es , s a ing ha o es s
al eady ha e a high le el o biodi e si y. In scena ios wi h exis ing hedges
(sc10-sc12), he mos equen ly chosen op ions we e hose enhancing
habi a connec i i y and minimising dis u bance o ield wo k.
I is wo h no ing ha many a me s assumed scena io cha ac e is-
ics ha we e no p o ided, such as a no h–sou h alignmen . Conse-
quen ly, se e al Es onian a me s placed he hedge ows o he no h o
a oid shading c ops, while Spanish a me s posi ioned hem o he sou h
o shield c ops om p e ailing ho winds. Al hough hese choices we e
excluded om he analysis, hey o e insigh s in o addi ional in lu-
encing ac o s.
3.2. The ole o ield and landscape cha ac e is ics
The scena io cha ac e is ics had a signi ican impac on he a me s’
choices (Table 3). In sc01, he p esence o a slope had a p onounced
e ec , wi h an excep ionally high odds a io obse ed o wid hwise
op ions. In he scena ios ela ed o soil quali y, he e was a no able shi
in choices, as a me s p e e ed o place measu es in a eas wi h lowe
soil quali y.
An inc ease in ield size was clea ly associa ed wi h highe odds o
subdi iding he ield. Al hough he esul s o he wind scena io should
be in e p e ed wi h cau ion, he e was a s ong indica ion ha he odds
o implemen ing hedges wid hwise, pe pendicula o he wind di ec ion,
we e subs an ially highe compa ed o he baseline scena io.
The p esence o o es s and oads was ound o inc ease he odds o
c ea ing habi a s pa allel o hese ea u es. Fo example, he odds o
plan ing a hedge ow along he leng hwise edge we e 82.3 imes highe
in he p esence o a oad. While an exis ing hedge along he sho e edge
o he ield (sc10) did no ha e an signi ican e ec , he oppo uni y o
connec wo hedges on neighbou ing ields (sc11) inc eased he odds o
choosing he wid hwise op ion by a ac o o 3.7.
The inal scena io (sc12) di e ed om he o he s in ha a me s
we e gi en he addi ional op ion o implemen ing a con inuous hedge-
ow along bo h edges o connec exis ing habi a s. In his scena io, he
odds o selec ing he op ion ha de ia ed om he s anda d op ions A
and B inc eased by a ac o o 11.7, despi e he highe e o equi ed.
3.3. The impac o a me and a m a ibu es
In he baseline scena io (sc00), he egion o he a m, a me s’a -
i udes owa ds biodi e si y, he sample g oup (hedge ow o lowe
s ip), and he maximum wo king wid h signi ican ly in luenced he
choices (Table 4). Speci ically, op ing o al e na i es o he a ou ed
choice (leng hwise a he edge) was associa ed wi h highe odds o
a me s om Eas e n (Region 2) and Sou he n Eu ope (Region 3),
highligh ing egional di e ences in p e e ences. Fa me s wi h s ong
biodi e si y a i udes had lowe odds o choosing op ion B (wid hwise a
he edge) bu highe odds o in- ield op ion C (leng hwise in- ield).
Region had he s onges o e all e ec and was signi ican ac oss
mos scena ios (Table 5). Fa me s in eas e n and sou he n Eu ope we e
mo e inclined o spli plo s. Fo example, in Region 3, he odds o
choosing op ion D (in- ield wid hwise) in he slope scena io sc01 we e
7.7 imes highe han in Region 1 (see Table S2 in he supplemen a y
ma e ial). These a me s we e also less keen on placing measu es on
low-quali y land bu had lowe odds o selec ing op ions o habi a
connec i i y. In pa icula , he odds o choosing he connec i i y op ion
E in sc12 we e 89 % lowe o Region 2 compa ed o Region 1 (Table S3).
Ins ead, a me s in Regions 2 and 3 p e e ed he g ea es possible dis-
ance om exis ing landscape elemen s, aiming o dis ibu e he mea-
su es mo e e enly ac oss he landscape.
Fu he mo e, a i udes owa ds biodi e si y had a signi ican in lu-
ence on he choices in he majo i y o scena ios (Table 5). Fa me s wi h
s ong a i udes had gene ally highe odds o choosing in- ield op ions
and conside ably lowe odds o selec ing op ions adjacen o exis ing
habi a s (sc07, sc08, sc10, sc11). These a me s also demons a ed a
g ea e willingness o dedica e land o biodi e si y in e en ions o
acili a e habi a connec i i y (sc12).
The esponses o he open-ended ques ions e ealed a numbe o
dis inc a ionales o he alue placed on he in e en ions. Flowe
s ips we e alued o soil e ili y, wildli e suppo , and aes he ics,
whe eas hedge ows we e app ecia ed o hei e ec on wind egimes
and ba ie unc ions. In he baseline scena io, he odds o placing a
hedge ow wi hin he ield we e 67 % lowe compa ed o lowe s ips
(Table 4). Addi ionally, hedge ows we e mos equen ly selec ed o be
a igh angles o exis ing habi a s such as o es s.
Mo eo e , he size o machine y, exp essed as he maximum wo king
wid h, exe ed a conside able in luence on he choices unde he base-
line condi ions. In p inciple, he e was a s onge p e e ence o
leng hwise op ions, e en wi hin he ield, when la ge machine y was
in ol ed. Howe e , his in luence was minimal in o he scena ios.
Con a y o ou ini ial expec a ions, he ype o land use (a able o
g assland) had a negligible impac on he decisions. Ne e heless, he
easons ci ed in he open- ex ield sugges ha hedges a e o en plan ed
as encing o o p o ec animals om he sun, pollu ion, o noise.
O ganic ce i ica ion appea ed o ha e only a mino e ec , and no sig-
ni ican in e ac ion wi h ield size o opog aphical elie was obse ed
o he egions in which he a me s wo ked.
4. Discussion
By explo ing a me s’spa ial p e e ences o biodi e si y measu es,
his s udy sheds ligh on how bo h ope a ional and socio-cul u al ac o s
in luence hei choices. The indings emphasise he necessi y o
designing conse a ion s a egies ha a e esponsi e o egional con-
ex s and a me p io i ies, he eby os e ing p ac icabili y and accep-
ance among a me s. They also open a enues o u u e esea ch o
in es iga e how spa ially in o med app oaches can enhance habi a
connec i i y while add essing mul iple ade-o s.
4.1. Discussion o esul s
P e ious esea ch has la gely concen a ed on he ole o ield, a m,
and a me cha ac e is ics on whe he ag i-en i onmen al measu es
(AEMs) a e adop ed. The p esen s udy ex ends his discussion by
examining whe e hese measu es a e mos likely o be placed, shi ing he
ocus om he a m o egional scale o he plo le el. Al hough he
espec i e esul s a e no di ec ly compa able, common pa e ns
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
5
Table 3
In luence o ield and landscape cha ac e is ics on a me s’p e e ences o alloca ing o linea wildli e habi a s a he ield scale (N=471).
Scena io Choice OR S d. e o z alue P (>|z|)
sc01 slope
H
L
H
L
A 0.09 0.27 −8.76 <0.001***
B10.05 0.28 8.34 <0.001***
C0.00 0.99 −10.59 <0.001***
D13033.37 0.88 10.80 <0.001***
sc02 soil quali y A 3.95 0.20 6.77 <0.001***
B
†
C 0.22 0.41 −3.70 <0.001***
D 4.10 0.82 1.73 0.084
sc03 soil quali y A 0.07 0.30 −9.03 <0.001***
B 16.27 0.31 9.12 <0.001***
C 0.00 0.91 −6.39 <0.001***
D
†
sc04 shape A 3.69 0.31 4.17 <0.001***
B 0.16 0.63 −2.88 0.004**
C 0.00 2.18 −6.96 <0.001***
D 6.36 2.18 0.85 0.395
sc05 size
x2
A 0.32 0.31 −3.76 <0.001***
B 1.90 0.58 1.11 0.268
C 109.33 1.64 2.86 0.004**
D 6813.31 1.77 4.98 <0.001***
sc06 wind A 0.10 0.33 −7.13 <0.001***
B
‡
11.59 0.23 10.44 <0.001***
C 0.00 4.26 −3.78 <0.001***
D 1.56 0.95 0.47 0.641
sc07 o es A 1.47 0.20 1.89 0.059
B 0.68 0.20 −1.89 0.059
sc08 o es A 0.15 0.22 −8.38 <0.001***
B 6.55 0.22 8.38 <0.001***
sc09 oad A 82.27 1.09 4.03 <0.001***
B 0.01 1.10 −4.03 <0.001***
sc10 exis ing hedges A 1.57 0.28 1.62 0.105
B 0.64 0.28 −1.62 0.105
sc11 exis ing hedges A 0.27 0.27 −4.76 <0.001***
B 3.69 0.27 4.76 <0.001***
sc12 exis ing hedges A 0.18 0.29 −5.97 <0.001***
B 0.45 0.54 −1.49 0.135
E11.69 0.40 6.09 <0.001***
Signi . codes: 0 [***] 0.001 [**] 0.01 [*] 0.05 [.] 0.1 [ ] 1. The columns show he s anda dised choice op ions, odds a ios o selec ing an op ion in he scena io
compa ed o he baseline, and signi icance le els ob ained om he GLMMs. Sc12 had he addi ional op ion o place a hedge ow a bo h he long and sho edges
(choice E) o connec exis ing hedge ows.
†
The models sc02_B and sc03_D we e uniden i iable due o a lack o sha ed IDs, indica ing ha he da a a iabili y was insu icien o es ima e he model pa ame e s.
‡
The model sc06_B encoun e ed bounda y singula i s. Applying a log-gamma p io (
α
=2, λ=0.001) o he andom e ec s did no esol e he issue, which
indica es insu icien da a o eliable es ima ion.
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
6
eme ge, indica ing unde lying simila i ies in he ac o s d i ing hese
decisions.
The esul s imply ha a me s’decisions ega ding he placemen o
biodi e si y measu es a e an exp ession o hei e o s o balance p o-
duc i i y, ope a ional e iciency, and he p omo ion o biodi e si y
h ough enhanced landscape connec i i y. The speci ic landscape
cha ac e is ics, machine y size, egional cul u al ac o s, and indi idual
a i udes owa ds biodi e si y s ongly in luence hese decisions. Key
ield pa ame e s, such as slope, soil quali y, ield size, and p oximi y o
exis ing landscape ea u es, u he guide he spa ial alloca ion a he
ield scale.
The e is a no able endency o a me s o a ou leng hwise place-
men o biodi e si y measu es, especially when using la ge machine y.
This p e e ence is consis en wi h he ypical a angemen o ac o
amlines in ec angula ields (Mede le &Be nha d , 2017), whe e
wid hwise obs acles impede he u ning o machine y. Leng hwise
managemen is no only ag onomically e icien bu also o e s ecolog-
ical ad an ages, as i educes headlands, whe e soil dis u bance and
compac ion nega i ely a ec in e eb a e abundance (Ca lesso e al.,
2022). Howe e , he p e e ence o placing wildli e habi a s leng hwise
along amlines appea s o be coun e balanced by ield-speci ic cha -
ac e is ics in o he scena ios.
Fa me s consis en ly a ge a eas o lowe soil quali y o AEM up-
ake, a pa e n suppo ed by ea lie s udies (Bo so o e al., 2008; Hynes
&Ga ey, 2009; Paulus e al., 2022; Rois-Díaz e al., 2018; Russi e al.,
2016; Wool e al., 2023). Addi ionally, p e ious esea ch has shown ha
conce ns abou soil e osion a e a s ong mo i a o o AEM adop ion
(F üh-Mülle e al., 2019; an He zele e al., 2013). The p esen s udy
Table 4
In luence o selec ed a iables on a me s’p e e ences o alloca ing linea wildli e habi a s in he baseline scena io (n =417).
sc00 [baseline]
Va iable Choice OR S d. e o z alue P (>|z|) P (>Chisq)
(In e cep ) B 2.90 0.65 1.65 0.100
C 0.01 1.02 −4.19 <0.001***
ATT_BIODIV B 0.76 0.07 −4.20 <0.001*** <0.001***
C 1.22 0.10 2.00 0.046*
FIELDSIZE B 1.00 0.01 −0.26 0.795 0.946
C 1.00 0.01 −0.26 0.799
LU_GRASS B 1.26 0.28 0.81 0.415 0.711
C 1.00 0.01 −0.26 0.799
ORGANIC B 0.91 0.28 −0.32 0.749 0.215
C 1.64 0.31 1.59 0.111
REGION2 B 2.20 0.34 2.33 0.020* 0.003**
C 3.88 0.39 3.46 <0.001***
REGION3 B 1.30 0.33 0.81 0.419
C 2.66 0.42 2.35 0.019*
RELIEF_HILL B 1.46 0.27 1.39 0.165 0.217
C 1.52 0.31 1.35 0.176
SAMPLE_H B 0.75 0.26 −1.13 0.257 0.014*
C 0.44 0.30 −2.83 0.005**
WORKWIDTH B 0.98 0.02 −0.91 0.364 0.012**
C 1.06 0.02 2.68 0.007**
Re e ence choice: long edge . Signi . codes: 0 [***] 0.001 [**] 0.01 [*] 0.05 [.] 0.1 [ ] 1, based on he Wald es (P (>|z|)) o each choice and he likelihood a io
es o he a iables (P (>Chisq)).
Table 5
Resul o likelihood a io es s es ima ing he in luence o pa ame e s in he scena io-speci ic models (N=471).
ATT_BIODIV FIELDSIZE LU ORGANIC REGION RELIEF SAMPLE WORKWIDTH
sc00 <0.001*** 0.946 0.711 0.215 0.003** 0.217 0.014* 0.012**
sc01 <0.001*** 0.110 0.185 0.122 <0.001*** 0.450 0.022* 0.139
sc02 0.068 0.167 0.633 0.564 <0.001*** 0.717 0.044* 0.090
sc03 0.003** 0.231 0.361 0.286 <0.001*** 0.995 0.044* 0.061
sc04 <0.001*** 0.180 0.002** 0.022* 0.048* 0.214 0.445
sc05 0.002** 0.730 0.637 0.421 0.629 0.985 0.073
sc06 0.884 0.644 0.873 0.127 0.007** 0.206 0.199
sc07 <0.001*** 0.479 0.425 <0.001*** <0.001*** 0.832 <0.001*** 0.574
sc08 0.003** 0.173 0.963 0.007** 0.007** 0.677 0.018* 0.606
sc09 0.036* 0.289 0.958 0.413 0.011* 0.700 0.606
sc10 0.008** 0.328 0.995 0.827 0.063 0.186 0.899
sc11 0.130 0.794 0.696 0.660 0.003** 0.123 0.374
sc12 <0.001*** 0.759 0.239 0.243 0.018* 0.621 0.092
P (>Chisq). Signi . codes: 0 [***] 0.001 [**] 0.01 [*] 0.05 [.] 0.1 [ ] 1.
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
7
unde sco es he ele ance o slopes and he associa ed isk o soil
e osion, which a e impo an ac o s in he spa ial alignmen o in-
e en ions a he plo le el.
A he ield scale, a me s’hedge ow placemen exempli ies how
hey weigh ope a ional ade-o s agains en i onmen al p io i ies,
pa icula ly in esponse o speci ic landscape ea u es. Fa me s
exp essed a clea p e e ence o loca ing in e en ions nea o es s,
which is in line wi h Paulus e al. (2022) who obse ed a g ea e like-
lihood o AEM adop ion nea woody ea u es. Fu he mo e, when gi en
he oppo uni y o link hedge ows o exis ing semi-na u al habi a s,
a me s adjus ed hei choices o he bene i o biodi e si y. This sug-
ges s a sensi i i y o habi a connec i i y and he en i onmen al impac
o measu es, wi h he p ospec o con inuing habi a s ou weighing he
desi e o minimise dis u bance in managemen p ac ices. Simila ly,
F üh-Mülle e al. (2019) epo ed a highe up ake o AEMs in Ge man
egions whe e habi a agmen a ion is a p essing conce n.
Alongside ield and landscape a ibu es, a me s’pe sonal cha ac-
e is ics a e cen al o hei decision-making. Nume ous s udies ha e
demons a ed ha posi i e a i udes owa ds biodi e si y and na u e in
gene al inc ease a me s’willingness o implemen biodi e si y- iendly
a ming measu es (see Klebl, Feind , &Pio , 2024a). This s udy adds
ha a me s wi h s ong p o-biodi e si y a i udes we e mo e inclined o
accep placemen op ions ha migh complica e plo cul i a ion, p i-
o i ising conse a ion objec i es o e ease o ope a ion.
The impac o egional speci ici ies as an o e a ching ac o u he
emphasises he pe sonal and con ex ual dimension o a me s’beha -
iou . The absence o collinea i y be ween he p edic o a iable REGION
and he cha ac e is ics o a ms and a me s indica es ha socio-cul u al
en i onmen s exe a conside able in luence on a me s’decisions
beyond a m s uc u al ac o s. This may be a ibu ed o egional habi s
and a ming adi ions, which signi ican ly shape indi idual ag icul u al
p ac ices (Pa lis e al., 2016; Rois-Díaz e al., 2018), illus a ing he
impo ance o unde s anding he socio-cul u al con ex s behind hese
choices.
While hese esul s o e aluable insigh s in o a me s’p e e ences,
i is c ucial o ecognise he po en ial dispa i y be ween s a ed p e e -
ences and ac ual beha iou . P e ious s udies sugges ha hypo he ical
scena ios, such as hose applied in his s udy, can o e es ima e he
pa icipan ’s willingness o ac (e.g., B owns one &Small, 2005; De
Co e e al., 2021; U ama &Hodge, 2006). None heless, he esul s
e eal clea choice pa e ns and highligh ade-o s ha a e ele an in
he design o a ge ed policies and in e en ions aimed a linking
wildli e habi a s.
4.2. Policy implica ions
The e is a pa icula need o spa ial planning a he landscape scale
o enhance habi a connec i i y (Pe’e e al., 2020; Pe’e e al., 2022).
Es ablishing a ne wo k o con inuing hedge ows is conside ed a iable
means o achie ing his in Eu opean landscapes and should be a ocus o
policy e o s (Moo house e al., 2014; S aley e al., 2023). One key in-
s umen o such policies is he p o ision o incen i es o landowne s
(Came on e al., 2022), which could be inco po a ed in o he Eu opean
Union’s Common Ag icul u al Policy (CAP).
So a , e idence poin s o he conclusion ha he CAP has no been
e ec i e in inc easing landscape connec i i y (Pa do e al., 2020).
Despi e he c i ical impo ance o spa ially coo dina ing linea land-
scape ea u es, he CAP S a egic Plans o he cu en legisla ion
(2023–27) do no sys ema ically add ess he spa ial componen in i s
ins umen s, i.e. he AEMs, now e e ed o as ag i-en i onmen -clima e
measu es (AECMs), Good Ag icul u al and En i onmen al Condi ions
(GAEC), o eco-schemes (Eu opean Commission e al., 2023). This
means ha a me s a e usually compensa ed o pe o ming ce ain
measu es bu , wi h some excep ions such as bu e s ips, he speci ic
placemen wi hin he ield is no aken in o accoun .
The e is po en ial o he widesp ead in oduc ion o an agglome -
a ion bonus ha ewa ds addi ional paymen s o AECMs when linking a
semi-na u al habi a o an exis ing one (Pa khu s e al., 2002), o
agglome a ion/ h eshold paymen s when a me s coope a e o
con ibu e o habi a connec i i y (D echsle e al., 2010; Nguyen e al.,
2022; W¨
a zold &D echsle , 2014). These ins umen s a e ela i ely easy
o implemen and o e p agma ic op ions. Ye , we also ad oca e o
mo e a ge ed app oaches de i ed om ecological ne wo k planning,
which may p o e mo e impac ul in connec ing speci ic habi a s o
complemen hese b oad schemes.
The spa ial coo dina ion and implemen a ion o planning-based
s a egies equi es ac i e collabo a ion be ween a ange o s ake-
holde s. Engaging a me s, go e nmen and public agencies, and con-
se a ion NGOs is cen al o de eloping a sha ed ision and e ec i ely
designing ecological ne wo ks (Keeley e al., 2018). Fos e ing coope a-
ion among a me s and communi ies can u he ad ance hese e o s
(McKenzie e al., 2013; Pe’e e al., 2022; Wes e ink e al., 2017), as
collabo a i e ini ia i es ha e been shown o subs an ially en ich a m-
land biodi e si y, no ably in e ms o bu e ly and bi d popula ions,
which is la gely a ibu ed o an inc ease in landscape connec i i y
(Meie e al., 2024).
Coo dina ed collabo a i e app oaches a e likely o deli e mo e
signi ican en i onmen al ou comes han indi idual ac ions, bu hey
a e associa ed wi h highe ansac ion cos s. Al hough hese ypes o
schemes may p o e e en mo e cos -e icien a la ge scales (Niemi
e al., 2024), hey mus comply wi h in e na ional subsidy egula ions
se by he Wo ld T ade O ganiza ion (WTO). In esponse o his, i has
been p oposed o edi ec inancial esou ces om es ablished
managemen -based schemes o collabo a i e ag i-en i onmen al
schemes (McKenzie e al., 2013). Mo eo e , policy amewo ks such
as he CAP ha e been c i icised o ailing o add ess s uc u al disin-
cen i es o collabo a ion, including land enu e a angemen s ha o en
con lic wi h he du a ion o AECMs (Le en on e al., 2017). These
misalignmen s a e o pa icula ele ance o pe manen landscape
ea u es such as hedge ows, whe e bo h enan s and landowne s need o
be in ol ed in conse a ion ag eemen s.
Beyond inancial compensa ion, he c ea ion o pla o ms o
knowledge exchange and coo dina ion be ween a me s, biodi e si y
ad iso s, and policy designe s is c ucial o e alua ing egional socio-
cul u al condi ions and na iga ing a m ope a ional and ecological ob-
jec i es. Le en on e al. (2019) ad oca e o go e nance sys ems aligned
wi h ecological scales ha complemen exis ing s uc u es. This would
en ail he o ma ion o a landscape-scale decision-making o um
comp ising di e se s akeholde s o collabo a i ely de elop conse a ion
plans o a de ined landscape. Howe e , he au ho s acknowledge ha
hese sys ems would equi e a undamen al eo ganisa ion o cu en
powe s uc u es and esponsibili ies.
Such pla o ms can also play a c i ical ole in ampli ying a i udes
owa ds biodi e si y ha ela e o a me s’ alues. The esul s o his
s udy indica e ha en i onmen al alues and he concep o landscape
connec i i y esona e wi h many a me s ac oss di e en egions, a m
ypes, and a m sizes, as shown by hei choices e lec ing conce ns
abou he b oade ecological impac o plo -scale in e en ions. We
he e o e ecommend g ea e in es men s in communica ing he alue
o a me s’po en ial o con ibu e o ecosys em heal h and he na u al
en i onmen in gene al. While con eying knowledge o ecosys em se -
ices de i ed om biodi e si y in e en ions is impo an , p e ious
esea ch sugges s ha add essing a me s’connec ions o he land, o
ins ance by emphasising he egion-speci ic cul u al meaning o
hedge ows, can be mo e e ec i e in secu ing long- e m commi men o
biodi e si y-enhancing measu es (Klebl e al., 2024b).
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
8
Fo s a egies a ge ing ecological ne wo ks o be success ul in e ms
o bo h biodi e si y conse a ion and inancial e iciency, Mossman
e al. (2015) speci y he need o su icien in o ma ion on he dis i-
bu ion o axa. This is pa icula ly pe inen in he con ex o ex-an e
assessmen s designed o iden i y he mos e icien way o connec
habi a s and o ensu e ha co ido s con ibu e o he expansion o
a ge species (Beie e al., 2011; B odie e al., 2016). Al hough p io i-
ising he sho es dis ance be ween agmen ed habi a s as he mos
cos -e ec i e solu ion is a plausible s a egy, we emphasise he consid-
e a ion ha he ue cos s o implemen a ion include compensa ion
paymen s o landowne s o −manage s, as also men ioned by Beie e al.
(2011) and Mossman e al. (2015).
Unde s anding landowne s’mo i a ions and p e e ences is he e-
o e c ucial o cos -e ec i eness and p ac icali y o conse a ion
s a egies. This s ems om he assump ion ha he g ea e he de ia-
ion o a planned ac i i y om a me s’p e e ences, he highe he
inancial compensa ion equi ed o encou age pa icipa ion. In his
way, he insigh s o he s udy could help o es ima e he cos s o
achie ing connec i i y. Linking woodland on a hill o a o es in a
alley, o example, may be mo e cos ly han es ablishing co ido s
pa allel o con ou lines due o e osion conce ns. Fu he mo e, co i-
do s along oads a e likely o be a low-cos solu ion, bu po en ial
ade-o s such as pollu ion, noise, and inc eased oadkill isks mus be
ca e ully weighed.
The in eg a ion o p edic ed beha iou al ou comes o s akeholde s
in o biogeog aphical and spa ial ecological models can be a powe ul
ool o iden i ying leas -e o s a egies and imp o ing ine-g ained
egional connec i i y maps (Beie e al., 2011). Such me hods can
enable p ac i ione s o a ge s akeholde s who a e mos likely o
implemen wildli e co ido s on hei land. Howe e , Be gs en and
Ze e be g (2013) highligh ed he lack o a sys ema ic app oach in
planning ecological ne wo ks. Ou esul s could p o ide one componen
in de eloping hese s a egies. To his end, we p opose o inco po a e
he pa ame e s ou lined in Table 6 in o spa ial ne wo k planning.
4.3. Resea ch ou look
The p ac ical ele ance o he s udy o ecological ne wo k planning
e lec s he need o esea ch-d i en analy ical ounda ions o in o m
decision-making and implemen a ion. Fo his pu pose, app oaches such
as agen -based modelling ha e been ecommended o simula e animal
mo emen and beha iou a he human-en i onmen in e ace, wi h
in e disciplina y app oaches being mos e ec i e (McLane e al., 2011).
By concep ualising a me s and biodi e si y as dis inc agen s, models
can es ima e he impac o di e en policy s a egies on a me beha -
iou and biodi e si y ou come (e.g., Dai e al., 2020; Djenon in e al.,
2022; Valbuena e al., 2010), and can be pa ame e ised wi h he
a iables iden i ied in his s udy.
A po en ial key ques ion o be add essed h ough such models is
which spa ial a angemen s o biodi e si y co ido s enhance habi a
connec i i y a leas cos while mee ing a me s’ope a ional needs. This
could in ol e a mul i-s age app oach: disc e e choice expe imen s o
quan i y he inancial compensa ion equi ed o implemen di e en
co ido scena ios and spa ial modelling o simula e cos -e ec i e
co ido con igu a ions wi hin a sample egion.
Beyond e alua ing cos -e ec i e a angemen s, es ima ing he
ecological impac o di e en con igu a ions is essen ial o unde -
s anding ade-o s be ween economic cos s and ecological e ec i eness,
hus helping o de e mine he mos iable op ions o habi a connec-
i i y. The indings o his s udy can s eng hen bo h newly de eloped
and exis ing models o ecological ne wo k planning, such as he Pa e o-
based app oach p oposed by G oo e al. (2010), which balances
ecological cohe ence, landscape cha ac e , and implemen a ion and
main enance cos s. In eg a ing social ac o s wi h ecological indica o s
would imp o e he p edic i e powe o hese models o assessing ade-
o s and ecological ou comes.
While hedge ows, pe ennial wild lowe s ips, and o he wildli e
co ido s a e widely ecognised o p omo ing biodi e si y, wa e
e en ion, pollina ion, soil p o ec ion, and soil ca bon seques a ion (e.
g., Holden e al., 2019; K a schme e al., 2024; Mon gome y e al.,
2020; Su e e al., 2018; Van Voo en e al., 2017), hey may ac as
ec o s o he sp ead o diseases and in asi e species (Mon gome y
e al., 2020), demanding ca e ul planning and managemen (Wilke son,
2014). Howe e , in ligh o he ongoing decline o exis ing co ido s and
he g owing agmen a ion o habi a s in Eu ope (A naiz-Schmi z e al.,
2018; EEA e al., 2011; Van Den Be ge e al., 2019), i emains ques-
ionable whe he hese isks ou weigh he subs an ial bene i s o ( e)
connec ing agmen ed habi a s. Ne e heless, inco po a ing ele an
isks in o modelling app oaches is ecommended o ensu e balanced and
eliable conse a ion planning.
Such e ined models may se e as he basis o he de elopmen o
p ac ical ools o suppo decision-making by ex ension se ices, go -
e nmen agencies, and p ac i ione s. To u he ex end hei impac ,
u u e esea ch could in es iga e which speci ic policy incen i es deli e
he g ea es en i onmen al bene i s a he lowes cos . This has he po-
en ial o in o m he design o policy ini ia i es ha a e e ec i e and
esponsi e o he needs o a me s.
5. Conclusions
This s udy shows ha a me s’decisions on he spa ial dis ibu ion o
biodi e si y measu es a e shaped by hei in insic mo i a ions and
p ac ical conside a ions. While a me s seek o op imise p oduc i i y
and minimise dis u bance o ield wo k, hey also place impo ance on
habi a connec i i y. The signi ican impac o egional ac o s and local
landscape cha ac e is ics on hei decisions unde sco es he need o
egion-speci ic conse a ion plans. Unde s anding a me s’p io i ies
and ailo ing e o s o speci ic socio-cul u al p ac ices and egional
condi ions can inc ease hei accep ance and e ec i eness.
The e is an oppo uni y o policy schemes, such as hose unde he
CAP, o be e in eg a e in e en ions ha con ibu e o habi a con-
nec i i y. Incen i es, including he agglome a ion bonus, can encou age
a me s o link habi a s, bu b oade , ecologically g ounded, and locally
adap ed solu ions a e conside ed o be mo e ui ul. I is he e o e
ecommended o policy designe s o s a egically add ess he spa ial
aspec s o conse a ion measu es o imp o e landscape connec i i y.
E ec i e p ac ical conse a ion elies on ac i e s akeholde
engagemen and collabo a ion among a me s. Collabo a i e ini ia i es
would bene i om egion-speci ic ecological guidance o en ich
Table 6
Pa ame e s p oposed o conside a ion when simula ing a me s’decisions o
alloca e linea wildli e habi a s.
Pa ame e P e e ence Mo i a ion
−leng hwise edge ease o managemen
slope wid hwise edge educ ion o uno and e osion
soil quali y low-quali y land leas impac on yield
wind egime windwa d edge c op p o ec ion
ca dinal
di ec ions
no he n/sou he n edge clima e- and c op-speci ic
shading
landscape
s uc u e
connec ion o habi a s/
along oads, o es s, ields
biodi e si y enhancemen /
p o ec ion om pollu ion and
wildli e
egional habi a s
and adi ions
egion-speci ic ag icul u al and aes he ic
p e e ences
F. Klebl e al. Landscape and U ban Planning 258 (2025) 105325
9
