Ci a ion:
Subiza, M. & Vozmediano, L. & San Juan, C. (2020). G een and blue se ings as p o ide s o men al
heal h ecosys em se ices: Compa ing u ban beaches and pa ks and building a p edic i e
model o psychological es o a ion. Landscape and U ban Planning. 204.
h ps://doi.o g/10.1016/j.landu bplan.2020.103926
Documen e sion:
A icle: Accep ed e sion
This is he accep ed manusc ip o he a icle ha appea ed in inal o m in Landscape and U ban Planning 204 : (2020) //
A icle ID 103926, which has been published in inal o m a h ps://doi.o g/10.1016/j.landu bplan.2020.103926.
© 2020 Else ie unde CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
G een and blue se ings as p o ide s o men al heal h ecosys em se ices: Compa ing u ban
beaches and pa ks and building a p edic i e model o psychological es o a ion
Mikel Subiza-Pé ez, Lau a Vozmedian, Césa San Juan
Uni e si y o he Basque Coun y
ABSTRACT
U ban beaches and pa ks a e p o ide s o nume ous ecosys em se ices. In he cul u al sphe e, place bonding and
psychological es o a ion migh signi ican ly con ibu e o he heal h and well-being o ci izens. In his manusc ip ,
we p esen a s udy aimed o e alua e he ex en o which h ee u ban beaches and h ee u ban pa ks o e ed hese
ad an ages o a sample o use s (n = 429) in he ci y o Donos ia-San Sebas ián (Spain). A second aim was o build a
p edic i e model o es o a ion h ough bo h objec i e and subjec i e measu es. We assessed he design and
physical ea u es o he se ings using he Na u al En i onmen Sco ing Tool (NEST) and ga he ed a ange o
in o ma ion abou he use s ia a pape & pencil ques ionnai e. The su ey included socio-demo- g aphics, ques ions
ega ding he equency and pa e ns o use, and ou di e en psycho-en i onmen al scales: Pe cei ed
Res o a i eness Scale (PRS), Place A achmen and Iden i ica ion Scale, and Res o a ion Ou come Scales (ROS). We
ound di e ences ega ding he p o ile o use s and he ac i i ies ca ied ou in each o he se ings. Use s o beaches
epo ed highe le els o a achmen , iden i ica ion, and expe ienced es o a ion han he pa icipan s su eyed in
u ban pa ks (p < .001). Reg ession analyses e ealed ha he main p edic o s o expe ienced es o a ion whe e
pe cei ed es o a i eness (β = 0.49), a achmen (β = 0.22), and iden i ica ion (β = 0.15), whe eas he
physical/design ea u es o he en i onmen and he ou ines o use made a negligible con ibu ion in his ega d.
The esul s o he eg ession analyses we e ex ended by conduc ing dominance and ela i e weigh analyses.
Keywo ds:
A en ion es o a ion S ess eco e y Place bonding Expe assessmen Field s udy
1. In oduc ion
Since he beginning o he cen u y, na u e and na u e-like en i onmen s and in as uc u es ha e been
alued o hei con ibu ion o se e al a eas ela ed o he li e and heal h o human beings. The Millennium Ecosys em
Assessmen epo (WHO, 2005) dis inguished be ween h ee ca ego ies o se ice ha such en i onmen s may o e :
1) he p o ision o esou ces necessa y o human li e (e.g., ood o esh wa e ), 2) he egula ion o na u al
phenomena such as empe a u e o he abso p ion o ainwa e , and, 3) he acili a ion o social and cul u al
in e ac ions and expe iences, all o which make a signi ican con ibu ion o human well-being (Fishe , Tu ne , &
Mo ling, 2009).
P e ious wo ks ha e highligh ed he mul i ude o se ices p o ided by g een and blue se ings and
in as uc u es in u ban con ex s, such as he educ ion o hea and ai pollu ion le els (Kabisch, an den Bosch, &
La o ezza, 2017). Mo eo e , he e is an inc easing in e es in hei ole in he mi iga ion and adap a ion o clima e
change h ough, o ex- ample, he educ ion o ene gy expendi u e o he mal egula ion in buildings (Demuze e e
al., 2014) and coping wi h ex eme wea he e en s (Voskamp & Van de Ven, 2015). In he social and cul u al sphe e,
mos o he s udied se ices a e ela ed o inc eased physical ac i i y – including mo e ac i e modes o anspo –
along wi h e- c ea ion, socializa ion, and he imp o emen o men al heal h (Demuze e e al., 2014; Kabisch e al.,
2017). Howe e , his ca ego y also comp ises aes he ic expe iences, a sense o place, and e en spi i uali y (Julian, Daly,
& Wea e , 2018) and appea s o ha e been de eloped o a lesse ex en when compa ed wi h o he ypes o se ices,
which is p obably due o some o i s de ining ea u es (Dickinson & Hobbs, 2017; Dou, Zhen, De G oo , Du, & Yu, 2017)1.
Acco ding o some au ho s, aes he ic enjoymen is a ele an mo i e o isi ing u ban g een and blue spaces (Dou e
al., 2017) and has impo an im- plica ions o human well-being due o he psychological es o a ion ha is gene a ed
by such spaces (Subiza-Pé ez, Hau u, Ko pela, Haapala, & Leh ä i a, 2019; Ul ich, 1983). A he ci y le el, pa ks and
beaches a e ela i ely la ge, di e se and impac ul se ings and migh he e o e be good p o ide s o hese so o
se ices. In his con ex , e en hough ci ies usually ha e u ban pa ks o o es s, he p esence o beaches wi hin u ban
bounda ies is limi ed o coas al se lemen s. In he case o Spain, many impo an ci ies (e.g. Ba celona, Valencia, Cádiz)
ha e one o mo e beaches which a e equen ly used by ci izens and ou is s seeking o es ing, socializing o
p ac icing spo s.
1.1. Men al heal h ecosys em se ices: people–place bonding and es o a ion
In his epig aph we p esen wo ecosys em se ices ha could be classi ied as men al heal h se ices
(B a man e al., 2019), and ha o m pa o he cul u al g oup o se ices, namely place bonding and psy- chological
es o a ion. Bo h se s o se ices ha e been he subjec o inqui y o esea che s on ecosys em se ices in ecen yea s
(B yce e al., 2016). In he ollowing pa ag aphs, we de ine hese se ices and desc ibe hei main con ibu ions o
human heal h and well-being.
In he li e a u e on sense o place, we can dis inguish be ween wo ela ed bu independen cons uc s,
namely place a achmen and place iden i ica ion (He nandez, Hidalgo, & Ruiz, 2014). Place a achmen is an a ec i e
bond ha people es ablish wi h places o loca ions ha a e impo an o hem, such as he home, he neighbou hood
o hei a ou i e leisu e des ina ion (Hidalgo & He nández, 2001; Lewicka, 2011). The cu en li e a u e desc ibes
how he places o which one is a ached p o ide he pe son wi h a ple ho a o bene i s – o psychological se ices –
anging om eelings o amilia i y, oo edness and sel - es eem, o a sense o elaxa ion and posi i e mood (Scannell
& Gi o d, 2017b, 2017a). On he o he hand, place iden i ica ion is he pa o he sel ha includes he meanings,
hough s and alues ha de ine a pe son’s iden i y, and which a e closely ela ed o places (D osel is & Vignoles, 2010;
Uzzell, Pol, & Badenas, 2002; Vale a & Pol, 1994). When a pe son iden i ies wi h a gi en place, hey expe ience a sense
o belongingness and a sense o pe sonal i wi h his loca ion. These se ings con ibu e o pe sonal well-being h ough
he s eng hening o sel -es eem and sel -e icacy and he eeling o sel -con inui y (Lewicka, 2008; Vale a, 1996).
Following Dickinson and Hobbs (2017), e en hough a sense o place makes a ele an con ibu ion o well-being,
mo e wo k is needed wi hin he ecosys em se ices li e a u e in o de o ully unde s and all o i s implica ions.
Res o a ion is de ined as he se o p ocesses leading o he enewal o physiological, psychological and social
esou ces ha become deple ed when coping wi h he asks and demands o e e yday li e (Ha ig, 2004, 2017). The
main heo ies o psychological es o a ion a e A en ion Res o a ion Theo y (Kaplan & Kaplan, 1989) and S ess
Reco e y Theo y (Ul ich, 1981, 1993), which cha ac e ize es o a i e en i onmen s as non-demanding and
aes he ically pleasing se ings, usually wi h li ing/na u e elemen s ha os e a sense o disconnec ion om he
s esses o daily li e, whils acili a ing he p ac ice o ac i i ies ha a e enjoyed by he indi idual. The a ailable
e idence indica es ha con ac wi h es o a i e en i onmen s educes physiological dis- ess, imp o es cogni i e
pe o mance, and igge s a mo e posi i e mood s a e (Bowle , Buyung-Ali, Knigh , & Pullin, 2010; Gascon e al., 2015;
Haluza, Schönbaue , & Ce inka, 2014; McMahan & Es es, 2015; Ohly e al., 2016).
A he ci y le el, u ban pa ks and beaches could be good p o ide s o bo h ypes o men al heal h ecosys em
se ices. These spaces usually o e high le els o biodi e si y and scenic beau y whils se ing as e uges om he
nuances and hassles ela ed o e e yday li e (hus les, wo k demands, and esponsibili ies). People usually isi hese
se ings o elax and amuse hemsel es, o p ac ice spo o o he ac i i ies, and o socialize. All o hese expe iences
os e he de elopmen o a ach- men and iden i ica ion and migh g an es o a ion o he isi o s o hose se ings.
1.2. The ole o psychological bonds in psychological es o a ion
Psychological es o a ion has adi ionally been unde s ood as an e olu iona y esponse o pa icula
en i onmen al ea u es and con- igu a ions ha ha e been essen ial o human su i al h oughou his o y (Kaplan
& Kaplan, 1989; Ul ich, 1993; Ul ich e al., 1991). Hence, es o a ion has been po ayed as a bo om-up p ocess, a
psychological esponse elici ed by he p esence o ce ain en i onmen al ea u es and/o con igu a ions. Howe e ,
in he las decade, hese explana ions ha e been b ough in o ques ion (Haga, Halin, Holmg en, & Sö q is , 2016).
Wo ks by Joye and collabo a o s (Joye & de Block, 2011; Joye & Dewi e, 2018; Joye & an den Be g, 2011) ha e high-
ligh ed conce ns ega ding, o example, he speci ic ea u es ha heo e ically b ing abou es o a ion and he
gene al es o a i e expe iences in a ious g een se ings ha appea o be lacking in he epo s ound in he
li e a u e. Fo ins ance, es o a ion heo ies posi ed ha es o a ion would occu in he p esence o en i onmen al
ea u es e- la ed o he e olu ion o he species (e.g. elemen s p o iding shel e o ood). Howe e , he e is a ele an
amoun o published e idence on he es o a i e e ec s o g een oo s, g een pa ks o e en u ban squa es, se ings
ha do no p o ide wi h many o hose se ices o ad an ages (Ha ig, E ans, Jamne , Da is, & Gä ling, 2003; Lee,
Williams, Sa gen , Williams, & Johnson, 2015; Subiza-Pé ez, Vozmediano, & San Juan, 2020).
Ra cli e & Ko pela (2016, 2017) asse ed ha one o he main limi a ions o conside ing psychological
es o a ion as a uni e sal e olu iona y-based esponse igge ed by ce ain en i onmen al ea u es is he subsequen
neglec o pe sonal a iables ha could also accoun o his esponse and claimed o mo e s udies using a op- down
pe spec i e. Indeed, he wo ks o hese au ho s e lec a new end in es o a ion s udies ha has ecen ly eme ged,
an app oach ha conside s he ole o o he a iables and mechanisms. Resea che s ha e begun o analyse which
pe sonal and social a iables –among o he s – migh also be implica ed in he expe ience o es o a ion om a op-
down pe spec i e. Fo example, Haga e al. (2016) showed ha he pe cei ed es o a i e po en ial and he ac ual
es o a i e e ec o a gi en sound a ied depending on whe he i was p esen ed o he pa icipan s as a na u al o
indus ial sound. This indica es ha he meaning asc ibed o a s imulus could mode a e i s es o a i e po en ial.
Recen esea ch has also shown ha a achmen and iden i y ied o ce ain en i onmen s can boos o s eng hen
he es o a i e expe ience o hese and he associa ed salu ogenic ou comes (Knez & Eliasson, 2017; Knez, Sang,
Gunna sson, & Hedblom, 2018; Liu e al., 2020; Mena i, Subiza-Pé ez, Villalpando-Flo es, Vozmediano, & San Juan,
2019; Mo on, an de Bles, & Haslam, 2017; Wilkie & Clous on, 2015; Wilkie & S a idou, 2013; Ysseldyk, Haslam, &
Mo on, 2016). O special in e es o his s udy a e he esul s o wo p e ious wo ks ha seem o indica e he
exis ence o a s onge associa ion be ween place a achmen and es o a ion han be ween place iden i ica ion and
he la e (Mena i e al., 2019; Subiza-Pé ez, Vozmediano, & San Juan, 2017). These indings a e compa ible wi h
hose epo ed by Knez and colleagues, who ha e consis en ly shown ha in compa ison wi h cogni i e componen s,
he emo ional componen s o place bonding a e mo e closely ela ed o well-being when spending ime in na u al and
u ban g een se ings (Knez & Eliasson, 2017; Knez e al., 2018).
1.3. S udy aim and objec i es
Wi h his s udy, we aimed o e alua e wo pa icula men al heal h se ices ha may be expe ienced by he
use s o u ban beaches and pa ks: place bonding and psychological es o a ion. Ou goal was o es ablish he
magni ude o hose expe iences in hese se ings and e alua e he con ibu ion o a wide ange o objec i e,
beha iou al, and psychological a iables owa ds he expe ience o es o a ion. Gi en he p emises de eloped in
Epig aph 1.2., we we e also in igued by he possible ole o a achmen and iden i ica ion wi h hose se ings in he
expe ience o es o a ion. We collec ed da a on hese a iables using an objec i e assessmen o he se ings and by
conduc ing a ield su ey. This allowed us o ob ain in o ma ion abou he physical/design ea u es o he s udy
se ings and he ou ines and psychological expe iences o hei use s. We hypo hesized ha he objec i e
cha ac e is ics o he se ings, he pa e ns o use, and he psychological bonding o he place would be ela ed o he
es o a ion expe ienced when spending ime in ha pa icula place. Fu he , we also wan ed o cha ac e ize he use
p o iles, pa e ns o use, psychological es o a ion, and pe son- place bonding in he wo ypes o se ings, de ec ing
signi ican di e ences (i any) ha migh exis be ween hem.
2. Me hods
2.1. Pa icipan s and s udy se ings
A o al o 429 people pa icipa ed in he s udy, o which 219 indica ed hei gende as emale (51.05%). The mean
age o he pa icipan s was 40.72 yea s (SD = 17.82). The sample was ec ui ed om he use s o h ee u ban pa ks
(“g een se ings”; n = 215) and h ee u ban beaches (“blue se ings”; n = 214) in Donos ia-San Sebas ián, a medium
sized Eu opean ci y in he Basque Coun y ha is home o a ound 180.000 people and i is loca ed on he No he n
coas o Spain. I has been desc ibed elsewhe e as a bou geois ci y wi h a ma ked compac g id pa e n design
(Fe nández Cues a, 2011) which makes i iendly and walkable. Comme ce and ou ism – bo h a he na ional and
in e na ional le el (Ga cía-He nández, de la Calle-Vaque o, & Yube o, 2017) – a e he main ac i i ies o he ci y, which
also s ands ou o i s high le els o non-mo o ized mobili y (Oses, Rojí, Gu u xaga, & La au i, 2017). I has a
ema kable aes he ic po en ial, bo h in na u al and a chi ec u al e ms, which pa ially lies in i s ichness o open
na u al spaces such as p omenades along he coas , along wi h he i e ha c osses he ci y and beaches. I also boas s
se e al pa ks o a conside able size, which a e equen ly used by he esiden s o he ci y.
2.2. Ins umen s
We conduc ed he objec i e assessmen o he s udy se ings using he Na u al En i onmen Sco ing Tool (NEST;
Gidlow e al., 2018). This ool is composed o 47 i ems di ided in o eigh domains: Accessibili y, Rec ea ion acili ies,
Ameni ies, Aes he ics − na u al, Aes he ics – non- na u al, Signi ican na u al ea u es, Inci ili ies and Usabili y. In
addi ion o assigning a sco e o each domain, NEST allows o he calcula ion o an o e all sco e. We decided o use
NEST because i was a ecen ly de eloped ool, which was designed a e a comp ehensi e e iew o p e ious ools,
and allowed o calcula e an o e all quali y sco e sensi i e o each se ing ypology (u ban beaches and pa ks in his
s udy).
The ques ionnai e o he use s o he se ings was p omp ed by p e ious esea ch (Ca us e al., 2015;
La o ezza, Ca us, Sanesi, & Da ies, 2009; Subiza-Pé ez, Vozmediano, & San Juan, 2019b). I included ques ions
ega ding demog aphics (age and gende ) and use o he place (dis ance om esidence, weekly and mon hly
equency o use, leng h o use, and pe o med ac i i ies). We egis e ed nine di e en ac i i ies: walking, mee ing
iends and ela i es, pe o mance o physical ac i i y, eading, landscape con empla ion, sunba hing/en- joying he
sun, d inking/ea ing some hing, spending ime wi h dependan s (e.g., child en) and walking he dog. Pa icipan s
had o indica e whe he hey usually pe o m hose ac i i ies (Yes/No) in he se ing whe e hey we e in e iewed.
The second pa o he ques ionnai e con ained he Spanish sho e sion o he Pe cei ed Res o a i eness Scale
(PRS; Neg ín, He nández- Fe naud, Hess, & He nández, 2017; α = 0.79), wi h 5 i ems measu ing being away, ascina ion,
cohe ence, compa ibili y and scope (e.g. “This is a ascina ing place ha keeps my cu iosi y ali e and s ops me om ge -
ing bo ed”). I also included he Spanish e sion o he Res o a ion Ou come Scale (ROS-S; Subiza-Pé ez, Vozmediano,
& San Juan, 2017; α = 0.93; e.g. “I eel calme a e being he e), an 8-i em scale measu ing he main aspec s o a
es o a i e expe ience: elaxa ion and calmness, a en ion es o a ion, clea ing one’s hough s, and e lec ion. Finally,
he ques ionnai e also con ained he Place A achmen and Place Iden i ica ion Scale (Ruiz, He nández, & Hidalgo,
2011), in he e sion c ea ed by Subiza-Pé ez e al. (2017) consis ing o 9 i ems (6 o a - achmen – α = 0.91, e.g. ”I
would eg e no coming o his place” – and 3 o iden i ica ion – α = 0.96, e.g. “I eel ha I belong o his place). All
hese scales used a 0–5 Like esponse o ma .
2.3. P ocedu e
Th ee ained esea che s isi ed he six s udy si es a he same ime and assessed hem using NEST. Each audi o
analysed he se ings indi idually. Following his ask, he da a collec ion g oup isi ed he se ings a di e en imes
o he day bo h du ing he week and he weekend. Di e en ime slo s (11–13 h, 13–15 h and 16–18 h) we e selec ed
o ga he he maximum a iabili y ega ding use s and ac i i ies. Upon a i ing a he se ings, hey app oached people
he e and in o med hem abou he objec i es o he s udy. We de ined wo eligibili y c i e ia be o ehand: 1)
pa icipan s had o be equen use s o he place ( ou is s and i s /second- ime s we e no in i ed o pa icipa e) and
2) hey had o be aged a leas 18 yea s. Use s ha decided o ake pa we e gi en he ques ionnai e a ached o a
clipboa d and we e ully ins uc ed on how o comple e i . When inished, pa icipan s we e b ie ly deb ie ed and
kindly hanked o hei pa icipa ion.
Following his p ocedu e, he da a we e collec ed be ween June and July 2018.
2.4. Da a analyses
Fi s , a ings o he objec i e assessmen o blue and g een se ings we e compiled, calcula ing an a e age sco e
o each audi o and sub- domain. Wi h hese indi idual a e age sco es, we hen calcula ed he eliabili y o NEST using
he In aclass Co ela ion Coe icien (ICC). Second, we desc ip i ely assessed he p o ile o use s and he ac i i ies
hey pe o med in g een and blue se ings sepa a ely and, using chi- squa ed analyses and F- es s (MANOVA), we
checked whe he he e we e di e en pa e ns o usage. Thi d, we an a Welch’s F- es o compa e he pe cei ed
es o a i eness, place a achmen , place iden i ica ion, and expe ienced es o a ion be ween each se ing.
Finally, wi h he objec i e o building a p edic i e model o he es o a ion achie ed in he s udy se ings, we an
a hie a chical linea eg ession. Fo his analysis, aimed o he objec i e o unde s anding which a iables accoun o
es o a ion in u ban g een and blue se - ings, we used ROS-S as he ou come a iable and PRS as p edic o . We began
by conduc ing co ela ion analyses o de ec i any aspec s o he da a we ga he ed (e.g., objec i e measu es, gende , o
pe o med ac i i ies) we e signi ican ly associa ed wi h he es o a i e ou comes epo ed by pa icipan s. Va iables
signi ican ly ela ed o he ou come we e hen in oduced in he eg ession in he co esponding block; 1) objec i e
assessmen a iables, 2) demog aphics, 3) use o he se ing and ac i i ies and 4) psycho-en i onmen al a iables. Due
o he limi a ions o co ela ions and s anda dized eg ession coe icien s as indica o s o he con ibu ion o each
p edic o a iable in eg ession models (Budescu, 1993; Da ling on & Hayes, 2017; Johnson, 2000), we used wo SPSS
u ili ies o analyse he ole o each o he a iables main ained in he inal s ep o he hie a chical eg ession model.
Speci ically, we conduc ed a dominance analysis and es ima ed he ela i e weigh s o each p edic o by using he RLM
(Da ling on & Hayes, 2017) and MIMR-Raw (Lo enzo-Se a, Fe ando, & Chico, 2010) p og ams espec i ely.
Resul s
3.1. Objec i e assessmen o he s udy se ings
The esul s o he objec i e assessmen o he s udy se ings a e shown in Table 1. Reliabili y analyses
e ealed ha , ollowing Hallg en (2012), mos o he objec i e indexes measu ed wi h NEST showed good and
excellen pe o mance; Access (ICC = 0.77), Rec ea ional acili ies (ICC = 0.71), Ameni ies (ICC = 0.76), Aes he ics –
na u al (ICC = 0.98), Aes he ic – non-na u al (ICC = 0.63), Inci ili ies (ICC = 0.63), Signi ican na u al ea u es (ICC =
0.86) and Usabili y (ICC = 0.89). The o e all measu e also showed e y good in e nal consis ency (ICC = 0.89).
Inspec ion o he able e eals ha bo h ypes o se ings ecei ed simila a ings o ec ea ional acili ies,
ameni ies and aes he ic – non- na u al, inci ili ies. None heless, in compa ison wi h g een se ings, he blue se ings
we e a ed as being less accessible, ha ing lowe na u al aes he ic po en ial, and being mo e usable. Finally, he blue
and g een se ings ob ained simila global sco es.
3.2. Ac i i ies and use p o ile by ype o se ing
Pa icipan s li ed be ween 1 and 120 min (M = 20.32, SD = 19.44) walking dis ance om he s udy si es. They
usually isi ed he place whe e hey we e in e iewed 2.62 imes (SD = 1.97) a week and 10.40 imes (SD = 8.26) a
mon h and spen 107. 92 (SD = 96.45) minu es he e pe isi on a e age. Taking he sample as a whole, he majo i y
o he epo ed ac i i ies we e walking and eading (71.3% each) ollowed by landscape con empla ion (56.4%), sun-
ba hing (48%) and mee ing iends and ela i es (41%). Lowe equencies we e ound o exe cising (28.2%),
d inking/ea ing some hing (25.2%) o aking ca e o child en o dependan s (17.2%). The leas common ac i i y was
walking he dog (7.9%).
Wi h he aim o es ablishing whe he he e we e signi ican di e ences in use p o iles and usage pa e ns
be ween g een and blue se - ings, se e al analyses we e conduc ed. Pa icipan s ec ui ed in bo h se ings we e equal
in e ms o gende [χ2(2) = 1.07; p = .587]. A MANOVA (see Table 2) e ealed ha use s o blue se ings we e younge
han hei coun e pa s, li ed u he om he se ing in which hey we e su ey and used i mo e imes in a week and
a mon h. The esul s o his analysis also e ealed ha people spen a conside ably g ea e amoun o ime du ing hei
isi s o blue se ings han o g een se ings.
When analysing he dis ibu ion o ac i i ies acco ding o ype o se ing, we ound an unequal dis ibu ion o he
ac i i ies o walking [χ2(2) = 22.79; p < .001], mee ing iends and ela i es [χ2(2) = 16.67; p < .001], exe cising [χ2(2)
= 45.28; p < .001], con empla ing he landscape [χ2(2) = 7.74; p = .005], sun-ba hing [χ2(2) = 81.68; p < .001] and aking
ca e o child en and dependan s [χ2(2) = 4.27; p = .039]. These analyses indica ed ha people using he blue se ings
a e mo e likely o mee iends, exe cise, and sun- ba he whe eas use s o g een se ings show a g ea e endency o
walk, con empla e he landscape, and look a e o he people. Howe e , no s a is ically signi ican di e ences we e
ound o eading [χ2(2) = 0.87; p = .352], walking he dog [χ2(2) = 2.09; p = .149] and ea ing/d inking some hing [χ2(2)
= 3.07; p = .08].
3.3. Psychological expe ience o he se ings
Acco ding o he esul s o he analysis shown in Table 3, blue and g een se ings we e pe cei ed o be equally
es o a i e al hough he o me yielded g ea e a achmen and iden i ica ion sco es. As indica ed by he esponses o
he pa icipan s, es o a ion a es we e also highe in he beaches han in he pa ks. The size o he di e ences o
a achmen , iden i ica ion, and es o a ion was small (ηp2 = 0.03, 0.05 and 0.05 espec i ely).
3.4. P edic ion o psychological es o a ion in he s udy se ings
We an co ela ion analyses (see Table 4) o selec he a iables o he hie a chical eg ession model.
App oxima ely only hal o he a iables measu ed in he s udy we e ound o be signi ican ly associa ed (p < .05) wi h
expe ienced es o a ion. Males, in compa ison o women, epo ed lowe es o a i e ou comes. Only i e
physical/design ea u es we e signi ican ly ela ed o he ou come and his ela ionship was nega i e mos o he imes.
Va iables e lec ing use pa e ns and ac i i ies we e ha dly ela ed o ROS-S sco es, wi h only equency o isi s and
landscape con empla ion being below he s a is ical signi icance le el. Finally, psycho-en i onmen al a iables we e
mo e closely ela ed o such an ou come han he objec i e and use- ela ed a iables.
This in o ma ion was hen used o build a hie a chical eg ession
model (see Table 5) o p edic expe ienced es o a ion h ough he signi ican ly associa ed a iables (p < .05). Since
he a iables coming om he objec i e assessmen we e highly co ela ed – which would e en ually lead o
mul icollinea i y issues wi hin he model – we decided o in oduce he a iable wi h he highes associa ion wi h
he ou come (usabili y).
As explained in Sec ion 2.4, we inally conduc ed bo h dominance and ela i e weigh analyses o u he
assess he con ibu ion o he six signi ican p edic o s in S ep 3. Dominance indexes a e shown in Table 6. This
analysis e ealed ha he o de o dominance be ween p edic o s is he ollowing: pe cei ed es o a i eness > place
a ach- men > place iden i ica ion > usabili y > landscape con empla ion > equency o use (pe mon h).
Addi ionally, ela i e weigh analysis e ealed ha pe cei ed es o a i eness accoun ed o 53.2% o he o al
a iance explained by he model (58.5%) whe eas place a achmen and place iden i ica ion accoun ed o 23.6%
and 15.4% espec i ely. O mino ele ance we e usabili y (5.5%) he ac i i y o con empla ing he landscape
(1.6%), and he equency o use (0.8%).
4. Discussion
G een and blue se ings and in as uc u es wi hin ci ies a e key p o ide s o a numbe o ecosys em se ices. The
objec i e o his s udy was o e alua e he ex en o which u ban beaches and pa ks in he ci y o Donos ia-San
Sebas ián p o ided hei use s wi h place bonding and es o a ion ela ed ecosys em se ices and explo e he
associa ion be- ween a di e se se o a iables and he es o a ion expe ienced in such se ings. We ound ha
women and men used bo h ypes o se ings equally whe eas blue se ings we e mo e equen ly isi ed by young
people in compa ison wi h hei olde coun e pa s. Beach isi o s a - ended he beaches mo e o en and o longe
pe iods o ime and we e mo e likely o mee iends, as well as ake pa in exe cise and sunba hing. Con e sely,
isi o s o g een se ings we e mo e likely o walk, con empla e he landscape, and look a e o he people when
spending ime in ha loca ion. Acco ding o ou esul s, isi o s showed mode a e o high le els o psychological
a achmen o hese se ings and el ha hey could iden i y wi h hese places o a mode a e ex en . Mo eo e , hey
epo ed mode a e and high es o a i e expe iences he e. We ound ha beaches, in compa ison wi h pa ks,
igge ed g ea e le els o a achmen and iden i ica ion. Fu he , whils bo h ypes o se ings we e pe cei ed as
equally es o a i e, beaches also p o ided mo e in ense es o a i e expe iences. Ne e heless, he di e ences a e no
o la ge size (ηp2 < 0.05). Howe e , all hese conclusions a e no eadily gene alizable o he use s o such se ings as
a whole (wi hin he s udy ci y o ab oad) due o he da a collec ion and sampling s a egies implemen ed o his
s udy.
We also aimed o build a p edic i e model o psychological es o a ion using he physical/design a iables
o he se ings and he pa e ns o use displayed by he isi o s. Mo eo e , and in acco d wi h an eme ging line o
esea ch (Mena i e al., 2019; Mo on e al., 2017; Ra cli e & Ko pela, 2016, 2017; Wilkie & Clous on, 2015; Wilkie &
S a idou, 2013), we included a achmen and iden i ica ion wi h he se ings in o de o asce ain whe he hese
a iables ac ually make a meaning ul con ibu ion o he expe ience o es o a ion. The esul s o his analysis
e ealed ha , in o de o ele ance, pe cei ed es o a i eness o he se ing, a achmen and iden i ica ion, he
ac i i y o con empla ing he landscape and he usabili y o he se ing signi ican ly p edic ed such an ou come. Fi s ,
hese esul s encou age u he esea ch on op-down a iables in es o a ion since he psychological a iables we e
by a he mos impo an p edic o s in e ms o explained a iance. The ac ha he ex en o which a pe son ega ds
a place o be es o a i e is a ac o ha con ibu es owa ds he expe ience o es o a ion is compa ible wi h he
indings o p e ious esea ch (Haga e al., 2016; Ruiz, Pé ez, & He nández, 2013) and wi h he subjec i is pe spec i e
on landscape pe cep ion/expe ience ha posi s ha beau y – o , in his case, es o a ion – is in he look o he
beholde (He as-Esc ibano & de Pinedo-Ga cía, 2018). The ac ha place a achmen , compa ed wi h place
iden i ica ion, showed a s onge associa ion wi h es o a ion is in line wi h he esul s o p e ious s udies discussed
in he in oduc ion (Knez & Eliasson, 2017; Knez e al., 2018; Mena i e al., 2019; Subiza-Pé ez e al., 2017). B a man
e al. (2012) posi ed ha pa o he es o a i e bene i s o na u e and na u e-like se ings may come om he eeling
o “belonging o some- hing g ea e han onesel ”, which has been usually in e p e ed as connec ion wi h na u e
(Capaldi e al., 2014), which may o e bene i s simila o he inclusion in social g oups and iden i ies (B a man e al.,
2012; Maye , F an z, B uehlman-Senecal, & Dolli e , 2009). Howe e , we know om en i onmen al psychology
wo ks ha place a achmen and iden i ica ion may wo k also as ele an social and indi idual ca ego ies (D osel is
& Vignoles, 2010; Lewicka, 2011; Mo on e al., 2017; Vale a & Pol, 1994; Ysseldyk e al., 2016). The e o e, we could
expec ha some o he posi i e ewa ds o expending ime in he s udy se ings may come om a eeling o
connec edness wi h o he s and/o wi h pe sonally ele an con en s.
The model explained hal o he a iance (53%) o he ROS-S sco es. E en hough his ep esen s a
conside able p opo ion o he a iance, he e is undoub edly scope o iden i ying u he explana o y a iables.
Su p isingly, physical/design a iables we e only mildly associa ed wi h es o a ion sco es and explained no mo e
han 5% o he a iance. Mo eo e , access and aes he ic po en ial o na u al and non-na u al elemen s wi hin he
se ings we e nega i ely co ela ed wi h expe ienced es o a ion sco es, which does no ag ee wi h he main
p emises o es o a ion heo ies (Kaplan & Kaplan, 1989; Ul ich, 1983; Ul ich e al., 1991) and should be con i med
o ejec ed by u u e s udies. This unexpec ed esul could be ela ed o he use o The Na u al En i onmen Sco ing
Tool (Gidlow e al., 2018) o assess he s udy se ings. This ool was o iginally designed o measu e he quali y o
u ban na u al en i onmen s and lis ed se e al ac o s (sub-dimensions) ha migh encou age o discou age hei use
by ci izens. Al hough some sub-dimensions con e ge wi h ac o s highligh ed by es o a ion heo ies (Ameni ies &
Rec ea ional Facili ies – Compa ibili y; Aes he ics – Fascina ion), he ool lacks some o he a iables ha ha e
p e iously been linked o psychological es o a ion and landscape p e e ence (Kaplan & Kaplan, 1989; Lo hian, 2017;
Ul ich, 1993). Examples o hese include cohe ence, complexi y, mys e y, and biodi e si y. Thus, u u e s udies should
inco po a e hese a iables o ob ain a mo e ac- cu a e imp ession o he in luence o physical/design ea u es o
g een/ blue se ings on es o a ion. A his poin , i could be ad isable o assess he p esence o such quali ies no
only by coun ing he numbe o collec ing he p esence o a gi en elemen (e.g. ees, g ass, benches o oun ains),
bu also by including a subjec i e e alua ion (o he a e s) abou he p esence o absence o ea u es o a much mo e
abs ac na u e. In his ega d, a ecen ly published assessmen ool – he IE- PREU (Spanish ac onym o In en o y
o he Assessmen o he Res o a i e Po en ial o U ban Spaces) – which includes some o hese indica o s, migh
p o ide he basis o such u u e wo k (Subiza-Pé ez, Vozmediano, & San Juan, 2019).
In e es ingly, equency o use was a posi i e p edic o in he second s ep o he model, bu his became a
nega i e p edic o in he hi d s ep. This appea s o indica e ha a e con olling o place-bonding a iables2, he
mo e a pe son isi s a es o a i e en i onmen he lowe he ac ual eco e y expe ienced. In e ms o es o a ion
heo ies, wo po en ial explana ions can be pu o wa d o explain his obse a ion. Fi s , psychological deple ion o
a igue is a necessa y p e equisi e o es o a ion (Ha ig, 2017). Thus, a g ea e numbe o isi s could educe he
need o es o a ion in subsequen isi s because he e is less o be es o ed each ime. Second, i also seems plausible
ha mo e isi s could lead o habi ua ion o he en i onmen ha would he e o e igge less in e es , ascina ion,
o aes he ic enjoymen . Consequen ly, he educed imme sion and engagemen would hinde – a leas o some ex en
– he a es o eco e y and e eshmen .
In ou iew, he combina ion o objec i e and subjec i e measu es, he size o he sample and he numbe o
se ings assessed a e all s eng hs o his s udy. None heless, we also ecognize ha he e a e ce ain limi a ions ha
need o be conside ed. Fi s , due o he p ocedu e we ollowed o ec ui pa icipan s, he esul s migh be a ec ed
by a sel -selec ion bias and migh no be gene alizable o he use s o he s udy se ings as a whole. Second, as
p e iously discussed, he educed con ibu ion o physical/design a iables in he es o a ion model could be ela ed
o he ool used o e alua ing hese a iables, and he e o e u he s udies a e needed wi h al e na i e ools.
