Assessing the quality of agricultural landscape change with multiple dimensions

Land 2014, 3, 598-616; doi:10.3390/land3030598 land ISSN 2073-445X www.mdpi.com/jou nal/land/ A icle Assessing he Quali y o Ag icul u al Landscape Change wi h Mul iple Dimensions Eija Pou a 1,*, Ioanna G amma ikopoulou 1, Timo Hu me 2, Ka iina Soini 3 and Ma ja Uusi alo 4 1 Economic Resea ch, MTT Ag i ood Resea ch Finland, La oka anonkaa i 9, 00790 Helsinki, Finland; E-Mail: ioanna.g amm[email p o ec ed] 2 Plan P oduc ion Resea ch, MTT Ag i ood Resea ch Finland, Da um, Tie o ie, 31600 Jokioinen, Finland; E-Mail: imo.hu m[email p o ec ed] 3 Economic Resea ch, MTT Ag i ood Resea ch Finland, Labo a o ium Uu e ie 2 A, 31600 Jokioinen, Finland; E-Mail: ka iina.soini@m . i 4 Plan P oduc ion Resea ch, MTT Ag i ood Resea ch Finland, E elä an a 55, 96300 Ro aniemi, Finland; E-Mail: m[email p o ec ed] * Au ho o whom co espondence should be add essed; E-Mail: eija.pou a@m . i; Tel.: +358-29-531-7627. Recei ed: 19 Decembe 2013; in e ised o m: 18 June 2014 / Accep ed: 24 June 2014 / Published: 1 July 2014 Abs ac : Be e ecogni ion o public pe cep ions is called o in de eloping policies ha a ec landscape quali ies, such as ag i-en i onmen al policies. The p esen s udy ocused on he e alua ion o ypical ag icul u al landscapes in Finland. We u ilized and ope a ionalized he isual landscape quali y scales in oduced by T ei e al. (2006) and u he explo ed how hese scales can be applied in ci izen e alua ion o ag icul u al landscapes. F om landscape da a collec ed ia an In e ne su ey, we analysed whe he and how he a ibu es o ag icul u al landscapes we e linked o hei e alua ion. The esul s demons a ed ha isual concep s such as openness, na u alness, species ichness and he imp ession o being aken ca e o we e signi ican ly associa ed wi h six landscape a ibu es, i.e., g ain, ca le, bales, a mhouses, buses and dis u bances. A ela ionship be ween key landscape concep s and no ma i e e alua ion was ound. The no ma i e pleasan ness o he landscape also signi ican ly associa ed wi h indi idual landscape a ibu es and he socio-demog aphic cha ac e is ics o he pe cei e s. Keywo ds: landscape e alua ion; landscape a ibu es; landscape quali y scales OPEN ACCESS Land 2014, 3 599 1. In oduc ion The impo ance o he ag icul u al landscape o u al li elihoods, cul u e and he iden i y o u al esiden s is well ecognized [1,2]. Al hough u al policies, and ag i-en i onmen al schemes (AES) in pa icula , impose basic obliga ions on a me s o keep he landscape open and well managed, he implemen a ion o AES ac i i ies o en comp omises di e en aims in a ming [3,4]. Fu he mo e, AES do no accoun o pe cep ions o he ag icul u al landscape among he public. Recen policies ha e emphasised be e ecogni ion o public pe cep ions o landscape quali ies. Fo example, he Eu opean Landscape Con en ion [5] sugges s ha landscapes and hei changes should be iden i ied and assessed “by he in e es ed pa ies and he popula ion conce ned” (A icle 6). I also calls o p ocedu es o he pa icipa ion o he gene al public, local and egional au ho i ies, and o he pa ies wi h an in e es in he de ini ion and implemen a ion o landscape policies (A icle 5). Al hough esea ch has ocused on ci izen e alua ion o landscapes and landscape cha ac e is ics, his esea ch has e y o en conce ned ou ism, he i age o landscapes wi h a speci ic na u al alue. Howe e , he Eu opean Landscape Con en ion [5] and he Fa o Con en ion on he Value o Cul u al He i age o Socie y [6] emphasize e e yday landscapes. Acco ding o hese Con en ions, ag icul u al landscapes wi hou any speci ic alue, such as a high na u al alue o alue as adi ional bio opes, should a ac equal conce n. In pa icula , i is impo an o de ine hose ag icul u al landscape a ibu es ha a e changing in esponse o ag icul u al policy and de e mine he e ec s o hese changes on how he landscapes a e e alua ed. Some esea che s ha e aimed o cap u e he mul idimensionali y o landscape quali ies wi h a limi ed numbe o measu es [7–10]. These measu es a emp o combine a ious aspec s o he landscape ha a e assumed o a ec subjec i e landscape assessmen . Based on he li e a u e, T ei e al. [10] iden i ied nine key concep s ha desc ibe di e en cha ac e is ics o he landscape: s ewa dship, cohe ence, dis u bance, his o ici y, he isual scale, imageabili y, complexi y, na u alness and epheme a. These key isual concep s ocus on di e en aspec s o he landscape, and oge he esul in he holis ic expe ience o i s isual quali y. The concep s o T ei e al. [10] sugges ha some uni e sal ways exis o e alua e a landscape, e en hough cul u al and gene ic ac o s in luence ou pe cep ions, implying ha he concep s and con ex a e obse e dependen . T ei e al. [10] demons a ed wi h he help o ag a ian landscape pho og aphs he impo ance o he key concep s. They p o ided objec i e illus a ions o quali y bu did no use he concep s in measu ing ci izen pe cep ions o he landscape. The key isual concep s we e no e alua i e on a posi i e o nega i e scale, i.e., high o low quali y landscapes, bu as exp essed by T ei e al. [10], some o hese concep s may inc ease isual quali y. They s a ed ha empi ical esea ch is needed on he ela ionship be ween he concep s and landscape p e e ences. Al hough s uc u ed wi h key concep s, landscape quali y can be subjec i ely pe cei ed. Se enan and An op [11,12] measu ed indi idual pe cep ions o quali y, and demons a ed a co ela ion be ween some o he key isual concep s and he aes he ic quali y o he landscape in he case o Belgium. Howe e , only hal o hese concep s we e ound o be eliable p edic o s. Hence, mo e empi ical s udies a e needed on he ela ionship be ween key landscape concep s and no ma i e e alua ion o he landscape. E e yday ag icul u al landscapes a e mul idimensional, bo h in spa ial and empo al e ms [13]. Besides cul i a ed landscapes and semi-na u al bio opes (such as g azing lands), ag icul u al landscapes Land 2014, 3 600 a e comp ised o a ious elemen s, including wild na u e (plan s, ocks, wa e ), man-made elemen s such as buildings o di e en ages, oads and ansmission lines, and animals (wild and domes ica ed), as well as signs o a ming (bales, ences). P e ious li e a u e has demons a ed ha he p esence o a single landscape a ibu e may cause a ma ked change in he o e all subjec i e assessmen o he landscape s uc u e and quali y (e.g., [14–16]). Howe e , he ela ionship be ween landscape a ibu es, pa icula ly ag icul u al a ibu es, and key cha ac e is ics o landscape quali y, such as hose p esen ed by T ei e al. [10], is open o ques ion. Fu he mo e, esea ch has e ealed ha pe cep ions o expe iences o ag icul u al landscape quali y e lec a numbe o socio-economic, psychological and cul u al ac o s, such as age, p o ession, educa ion, and he sense o place o he pe cei e . A dis inc ion can he e o e usually be made, o example, be ween a me s, ci y dwelle s, expe s and conse a ionis s [17–20]. Pho og aphs ha e inc easingly been used o explo e landscape e alua ions (e.g., [18,19,21–25]). I pho og aphs a e used o e alua e he quali y o he landscape, he quali y o he pho og aphs hemsel es may a ec he e alua ion [26]. As a gued by Rose [27], an image has pa icula e ec s upon us, depending on i s con ex ual in o ma ion (ac ual and exp essi e con en s, colou , spa ial o ganiza ion and ligh ). Colou pho og aphs p o ide mo e in o ma ion on he landscape han black-and-whi e pho og aphs [28,29], bu hey can be mo e sensi i e o he di e ences caused by he wea he condi ions o he pe iod du ing he g owing season, o example. An expe imen by Shu lewo h [26] indica ed ha black-and-whi e pho og aphs end o induce mo e ex eme and mo e highly di e en ia ed esponses han colou pho og aphs, and ha he la e ela e mo e closely o ield esponses. In using pho og aphs o in es iga e landscape e alua ions, we also ook he oppo uni y o examine he e ec o he ype o pho og aph on e alua ions o a ious con ex s. The p esen s udy ocused on landscape e alua ions by ci izens based on pho og aphs ep esen ing i e ag icul u al landscapes in Finland wi h changing a ibu es. The i s objec i e was o use he indica o s sugges ed by T ei e al. [10] o measu e ci izen pe cep ions o landscape quali y om se e al landscape pho og aphs, as a limi ed amoun o space is a ailable in a su ey and esponden e o needs o be kep on mode a e le el. A second objec i e was o iden i y he ela ionships be ween he key concep s ( hose de eloped by T ei e al. [10]) measu able isual a ibu es o he ag icul u al landscape, and hei no ma i e (posi i e and nega i e) a ings. Fu he mo e, we compa ed landscape e alua ions be ween black-and-whi e and colou e sions o he same landscape pho og aphs o de e mine whe he measu es o key concep s a e sensi i e o he pho og aph ype. 2. Me hods 2.1. Su ey Design 2.1.1. Selec ion o Pho og aphs In he su ey, we illus a ed he landscape using pho og aphs. In landscape e alua ion s udies, he landscapes and hei a ibu es ha e mos o en been isualized wi h ae ial o landscape pho og aphs (e.g., [18,19,21–25]). He e, we also op ed o use pho og aphs a he han on-si e me hods, as hey allow mo e people o pa icipa e in he esea ch, make he esea ch less expensi e [19,30] and enable compa isons be ween di e en landscape ypes, since hey di ec he obse e ’s ocus o isual Land 2014, 3 601 quali ies ins ead o assessmen s based on o he senses [31]. A numbe o esea che s ha e epo ed high co ela ions be ween pho o-based and on-si e e alua ions o landscapes [32]. Pho og aphic isualiza ion is also an easily applicable me hod in landscape e alua ions ia In e ne ques ionnai es [33,34]. In ou In e ne su ey ques ionnai e, esponden s we e p esen ed pho og aphs o i e ag icul u al landscapes, wi h wo pho os o each landscape aken a di e en imes o he yea : in mid- and la e summe (Figu e 1). Fo hal o he esponden s, he pho os we e p esen ed in black and whi e, while colou pho os we e p esen ed o he o he hal in o de o examine he e ec o image ype on he assessmen s. Figu e 1. (1–10) Pho og aphs used in he su ey (colou e sions, la e summe on he le , mid-summe on he igh ). (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Land 2014, 3 602 The pho og aphs we e selec ed om an a chi e o 2950 images o ag icul u al landscapes in Finland om 13 di e en a eas ep esen a i ely a ound he coun y [35] aken in he yea s 2000, 2007 and 2010. 2.1.2. Selec ion o A ibu es P e ious esea ch p o ided u he guidelines o iden i ying he landscape a ibu es and selec ing he pho og aphs. Summa izing pas li e a u e guidelines [36], landscape cha ac e is ics may be g ouped in o wo main ca ego ies: he le el o managemen o ege a ion, and he s a us and condi ion o man-made elemen s in he landscape. Rega ding he cul i a ion s yle, Howley [37] and Howley e al. [38] concluded ha people p o ide highe e alua ions o adi ional, mo e ex ensi e a ming landscapes han mo e mode n, in ensi e a ming ones. In gene al, e alua ions a e lowe i landscapes consis o homogeneous monocul u es [18,39–41]. A iaza e al. [42] epo ed ha he isual quali y o he landscape inc eases as a unc ion o he pe cen age co e o ege a ion. Benjamin e al. [40] also demons a ed ha ege a ion ma e s, as abandoned a mland was ound o be he mos un a ou able landscape, ollowed by co n ields. Rech man [43] s udied he e ec o he c op ex u e and ound ha he p esence o ege a ion had a posi i e impac on p e e ences, bu also ha a mix u e o ield c ops and o cha ds was app ecia ed mo e han a homogeneous c op ex u e. P e ious s udies ha e shown ha man-made a ibu es ha e a powe ul e ec on e alua ions. Ru al buildings [44], pa icula ly a m buildings [45,46] and cul u al buildings [47,48], a e posi i ely ela ed o landscape e alua ions, especially when hese buildings a e old [41], adi ional [16,37,49] o well p ese ed [42]. Mo eo e , esponden s app o e o he p esence o isual di ide s, i.e., bu e zones ha con ain g assy o ee-co e ed a eas, hedge ows o e aces [15,45,50,51], and he p esence o g azing animals [37,46,49]. Beyond he p e ious li e a u e, ou aim was o ocus on hose a ibu es ha had ypically changed in he Finnish ag icul u al landscape due o policy and echnological changes: a dec ease in ca le p oduc ion in he sou he n pa o Finland, less equen u n-ou o ca le o pas u e, echnological changes in odde p oduc ion/hay s o age sys ems, he inc easing size o a m compounds and enla ged ield plo s leading o a dec ease in ield edges. The 10 pho og aphs om i e si es we e selec ed so ha isual cha ac e is ics in he ag icul u al landscapes could be iden i ied wi h landscape a ibu es. By applying he ideas om p e ious s udies o Finnish condi ions [35], some o he pho og aphs ep esen ed he g ain p oduc ion landscape, and some depic ed g ass p oduc ion o ca le a ming. Ca le a ming was also isible in wo pho og aphs as he p esence o ca le o silage bales. A one o he si es, he landscape a ia ion esul ed om he managemen o di ch edges by ei he emo ing o allowing bushes and small ees o g ow (pho os 1 and 2). The e was also a ia ion in he p esence o ag icul u al buildings, which we e included in i e o he pho og aphs. One o he pho og aphs (pho o 7) included a landscape dis u bance in he o m o a pile o b ushwood. The cha ac e is ics o he pho os could be con e ed in o nume ical in o ma ion on he landscape. The image con en could be exp essed in e ms o landscape cha ac e is ics ha we e coded in he da a as six dicho omous a iables: g ain/g ass, ca le, bales, a mhouses, bushes and dis u bances. These a iables a ied be ween he pho og aphs and si es. Land 2014, 3 603 A e iewing each pho og aph, he esponden s we e asked o e alua e he landscape. This was done by u ilizing he key concep s in oduced by T ei e al. [10], complemen ed wi h e alua ion o pleasan ness and biodi e si y. Using he concep s o T ei e al. as a s a ing poin , we applied he idea o seman ic di e en ial scales [52] and aimed o ind adjec i e pai s ha desc ibed and ope a ionalized each dimension in a way ha could be easily p esen ed o Finnish esponden s in he Finnish language. Consequen ly, he English ansla ions o he adjec i e pai s measu ed we e: pleasan –unpleasan , aken ca e o /main ained–no aken ca e o /no main ained (“s ewa dship”), consis en –di use (“cohe ence”), ha monious–dis u bing (“dis u bance”), in ol e his o y-only e lec he p esen (“his o ici y”); open–closed (“ isual scale”), o iginal– ypical (“imageabili y”), di e se–mono onous (“complexi y, di e si y”), na u al–human modi ied (“na u alness”), s able–changing (“epheme a”) and ich in species–poo in species (“biodi e si y”). The e alua ion was measu ed by asking esponden s o a e each pho og aph on a i e-poin scale be ween each wo d pai . These adjec i e pai s we e es ed in a pilo s udy, and as an adequa e sp ead o he da a o each measu e was ob ained and no indica ion o misunde s anding o he wo d pai s was obse ed, he adjec i e pai s we e accep ed in he inal su ey as such. In he ollowing, we use he i s adjec i e in each pai o desc ibe he dimension. 2.1.3. Measu es o Socio-Demog aphic P o ile P e ious esea ch has demons a ed ha he socio-demog aphic p o ile o he pe cei e , whe he a esiden o isi o , is associa ed wi h landscape pe cep ions. In pa icula , landscape e alua ions ha e been ound o be a ec ed by he pe cei e ’s educa ional le el [38,53–56] and gende [37,49,55]. Mo eo e , he leng h o esidence [57,58] and age o he pe cei e s [14,16,37,49,55,56] can be o impo ance, as hey ela e o p e ious expe iences and knowledge o he his o y o a landscape. The childhood en i onmen ( a m, u al o u ban) [37,49,51] signi ican ly a ec s people’s e alua ion o scenic beau y. People who a e mo e acquain ed wi h a landscape expe ience he landscape di e en ly and hus exp ess di e en p e e ences (e.g., [37]). Ru al dwelle s, i.e., people who li e in he coun yside bu whose li elihoods do no depend on ag icul u e, show di e en p e e ences compa ed o expe s o a me s [18]. Fu he mo e, a p o ession ela ed o he economic use o na u al esou ces has been ound o ha e an e ec on he e alua ion o landscape u iliza ion (e.g., [20,24,59–62]). Based on hese ideas om p e ious s udies, in addi ion o gende and age, he socio-demog aphic backg ound o he esponden s, such as educa ion, p o ession and income, and a iables ela ed o he li ing en i onmen , i.e., cu en and childhood li ing en i onmen , we e measu ed a he end o he su ey. 2.2. Da a An online In e ne su ey conduc ed in Ap il 2011 p o ided in o ma ion on he e alua ion o he ag icul u al landscape. The da a we e collec ed om he In e ne panel o a p i a e su ey company, Talous u kimus. The panel comp ised 30,000 esponden s who had olun ee ed o pa icipa e in he panel [63]. A e he pilo su ey o 100 people, a andom sample o 3016 esponden s was selec ed, and 800 people comple ed he su ey, esul ing in a esponse a e o 27%. The da a we e close o ep esen a i e o he gene al popula ion [64] ega ding gende (45% emales in he da a, 51% in he popula ion) and age (mean 48 yea s in he da a, 42 in he popula ion). The educa ional le el was somewha highe in he sample, as 32% o esponden s had a highe educa ion, while in he popula ion Land 2014, 3 604 hei sha e was 27%. The p opo ion o indi iduals wi h child en in he amily was clea ly highe in he sample (29%) han in he gene al popula ion (16%). The da a we e also somewha skewed o o e - ep esen people om o he pa s o he coun y han he mo e popula ed sou he n Finland (62% in he da a, 50% in he popula ion). 2.3. S a is ical Me hods Fi s , we analysed he ope a ionaliza ion o he scales de eloped by T ei e al. [10]. This was done wi h desc ip i e analysis and by calcula ing he Spea man ank co ela ions be ween measu es. The assump ion o mono onic ela ionships be ween a iables in Spea man co ela ion was in es iga ed wi h c oss abula ions and obse ed o be me in a clea majo i y o ela ionships. By co ela ing he measu es wi h he e alua ion o pleasan ness (pleasan –unpleasan ), we could also examine he associa ion o he measu es wi h no ma i e e alua ion. Nex , linea mixed models we e selec ed o explain he a ia ion in key isual concep s due o landscape a ibu es [65]. The e ec s o he i e si es and o each indi idual we e aken in o accoun as andom e ec s. All he landscape cha ac e is ics and socio-demog aphic a iables we e examined as po en ial explana o y a iables ( ixed e ec s) in hese models. The models we e buil so ha all he indi idually signi ican and consequen ly p omising a iables we e included in he model oge he . Then, he non-signi ican a iables we e dele ed om he model one by one based on he highes p- alue (Type 3 F- es s) in each educed model. The a iable elimina ion was con inued un il all he p- alues we e signi ican o e y close o signi ican . The model o no ma i e e alua ion is p esen ed in de ail, bu he 10 models o quali y scales a e only p esen ed wi h indica i e coe icien signs and signi icances o sa e space. Al hough no op imal o he si ua ion, as he esponse a iables we e measu ed on an o dinal i e-poin scale ins ead o as a con inuous measu e, he me hod was conside ed as he bes possible al e na i e, because i yields esul s in a use iendly o m. O he assump ions o he model we e es ed using g aphical me hods: esiduals we e plo ed agains he i ed alues and he no mali y o he esiduals was checked h ough quan ile-quan ile plo s, and he assump ions we e easonably me . The inal models o key isual concep s o he landscape we e cons uc ed so ha he po en ial explana o y a iables we e i s es ed indi idually wi h he andom e ec s in he model. Based on hese ini ial models, he signi ican a iables we e simul aneously added o he combined model, and he model was educed by elimina ing non-signi ican a iables one a a ime. The s a is ical modelling was pe o med using he MIXED p ocedu e o SAS 9.3 (SAS Ins i u e Inc., Ca y, NC, USA). The e ec o pho og aph ype (black and whi e s. colou ) was u he analysed by compa ing he means o e alua ions be ween ypes. Linea mixed models we e also used o analyse he e ec s o he pho og aph i sel , he pho og aph ype (black and whi e s. colou ) and hei in e ac ion in a ious concep s. In hese models, he e ec s o each indi idual we e conside ed as andom e ec s. Land 2014, 3 605 3. Resul s 3.1. E alua ing Measu es o Key Visual Concep s In Table 1, we p esen desc ip i e in o ma ion on he measu es o key isual concep s de eloped by T ei e al. [10], as well as on he measu es o pleasan ness and biodi e si y. The means o he measu es indica ed he pa icipan s’ pe cep ions o he openness o he ag icul u al landscape, as he openness quali y measu e was a ed highe in all pho og aphs han o he concep s. The means also exp essed he esponden s’ a he high pe cep ion ha he landscapes had been aken ca e o and ha hey we e ha monious and consis en . App oxima ely in he middle o he scales om 1 o 5 we e e alua ions o he na u alness, s abili y and o iginali y. No ma i ely, he landscapes we e pe cei ed as a he pleasan wi h a mean o 4.1 o all pho og aphs. The F- es showed ha he concep s we e able o e eal di e ences in landscapes be ween he pho og aphs, as he means di e ed signi ican ly o all concep s. Table 1 p esen s he lowes and highes e alua ions, showing ha he highes e alua ions, in pa icula , accumula e in indi idual pho og aphs (5). Table 1. Desc ip i e s a is ics o he key isual concep s measu ed on a scale om 1 (low) o 5 (high). The di e ences be ween pho og aphs we e s a is ically signi ican o all he measu ed concep s acco ding o he F- es (p < 0.001). Pho og aph 1 2 3 4 5 6 7 8 9 10 Means, Scale 1, ..., 5 Pleasan ness 4.41 4.29 4.20 4.26 4.31 4.12 3.81 3.96 3.63 3.89 Species ichness 3.39 3.18 3.47 3.06 3.88 3.30 3.88 3.06 3.21 3.43 Taken ca e o 4.15 4.29 3.93 4.42 3.02 4.22 2.34 4.17 3.37 3.87 Consis ency 3.77 3.92 3.63 3.87 3.47 3.62 3.08 3.47 3.26 3.30 Ha mony 4.07 4.00 3.81 3.95 3.89 3.56 3.27 3.31 3.33 3.39 In ol e his o y 3.50 3.30 3.55 3.32 3.74 3.18 3.68 3.02 3.53 3.37 Openness 3.97 4.24 3.76 3.78 3.66 3.68 3.77 3.83 3.73 3.82 O iginali y 2.93 2.75 2.93 2.85 3.09 2.68 2.92 2.73 2.76 2.79 Di e si y 3.58 3.20 3.55 3.34 3.80 3.31 3.48 3.27 3.04 3.40 Na u alness 2.63 2.38 2.65 2.16 3.70 2.38 3.84 2.22 2.70 2.41 S abili y 2.99 2.92 3.01 3.02 3.09 2.86 2.99 2.84 2.89 2.84 Bold highes alue; bold i alics lowes alue in each concep . SD—S anda d de ia ion. The co ela ions be ween key isual concep s (Table 2) indica ed ha nea ly all o hese, as well as he no ma i e measu es, we e signi ican ly associa ed wi h each o he . This implies ha as s a ed by T ei e al. [10], he concep s a e in e ela ed and wo k oge he o o m he o ali y o isual landscape. The a ia ion in co ela ions also indica ed ha some concep s a e mo e closely linked han o he s. The co ela ions we e especially weak be ween na u alness and o he measu es. Howe e , hey we e pa icula ly s ong be ween he measu e o di e si y and o he scales, and be ween ha mony and o he scales, as well as be ween ha mony and consis ency (0.534) and also be ween species ichness and di e si y (0.502). Land 2014, 3 606 F om he co ela ions wi h he e alua ion o pleasan ness (pleasan –unpleasan ), we we e also able o examine he associa ion o key concep s wi h he no ma i e posi i e–nega i e dimension. All o he measu es o key concep s co ela ed posi i ely and signi ican ly wi h pleasan ness. Rega ding some concep s, his was ob ious, as concep s such as ha mony can be easily in e p e ed o ep esen posi i e and nega i e e alua ions. The co ela ions wi h he no ma i e scale we e weakes o he concep o na u alness, which ne e heless had a signi ican posi i e associa ion wi h pleasan ness. Table 2. Spea man co ela ion coe icien s be ween 11 landscape e alua ion scales. Pleasan ness Taken Ca e o Consis ency Ha mony In ol e His o y Openness O iginali y Di e si y Na u alness S abili y Species Richness Pleasan ness 1 Taken ca e o 0.488 ** 1 Consis ency 0.451 ** 0.459 ** 1 Ha mony 0.645 ** 0.440 ** 0.534 ** 1 In ol e his o y 0.226 ** −0.054 ** 0.137 ** 0.193 ** 1 Openness 0.428 ** 0.312 ** 0.456 ** 0.437 ** 0.245 ** 1 O iginali y 0.239 ** 0.058 ** 0.166 ** 0.214 ** 0.241 ** 0.171 ** 1 Di e si y 0.450 ** 0.165 ** 0.252 ** 0.430 ** 0.306 ** 0.277 ** 0.337 ** 1 Na u alness 0.069 ** −0.356 ** −0.105 ** 0.060 ** 0.179 ** −0.048 ** 0.192 ** 0.253 ** 1 S abili y 0.165 ** 0.079 ** 0.128 ** 0.216 ** 0.063 ** 0.097 ** 0.059 ** 0.173 ** 0.269 ** 1 Species ichness 0.310 ** −0.012 0.144 ** 0.286 ** 0.308 ** 0.193 ** 0.264 ** 0.502 ** 0.388 ** 0.208 ** 1 ** Co ela ion is signi ican a he 0.01 le el (2- ailed). Co ela ions o e |0.4| wi h bold. 3.2. Modelling Landscape E alua ions The esul s o he linea mixed model o pleasan ness a e epo ed in Table 3. The model akes in o accoun he simul aneous e ec s o he ime o yea , pho og aph ype (black and whi e o colou ), landscape cha ac e is ics, socio-demog aphic a iables and andom e ec s o he si e and indi idual. Table 3 p o ides leas squa e means es ima es, which a e model-based means o each class ( o example, means o pleasan ness on a scale om 1 o 5 o black-and-whi e and colou ed pho os), o he wise assuming a e age indi idual and pho og aph cha ac e is ics. As can be seen in Table 3, i o he a iables we e held a he a e age le el, bushes by di ches imp o ed he e alua ion, while dis u bances (piles o b ushwood) educed he le el o pleasan ness. The p esence o ca le had a s a is ically signi ican , bu mino posi i e e ec on he gene al pleasan ness e alua ion. O he a ibu es (g ain, bales, a mhouses) had no signi ican e ec on pleasan ness. These non-signi icances may also ela e o co ela ions wi h o he mo e signi ican a iables in he model, o example in he case o a mhouses wi h he ca le, and in he case o bales wi h he ime o he yea , i.e., mid-summe . The pe cei ed pleasan ness o he pho og aphs nega i ely associa ed wi h hem being aken in la e summe , as he es ima ed means we e lowe wi h a somewha signi ican p- alue o 0.0498. Pleasan ness was a ec ed by he pho og aph ype, wi h black-and-whi e landscapes ecei ing sligh ly highe e alua ions. The model also e ealed a signi ican in e ac ion be ween pho og aph ype and he season. In mid-summe pho og aphs, he colou ed e sions we e pe cei ed as mo e pleasan han black and whi e, bu he e ec was opposi e in la e summe pho og aphs. Land 2014, 3 613 Au ho Con ibu ions Eija Pou a pa icipa ed in planning he su ey, analyzing Tables 1, 2 and 5, w i ing he esul s sec ion and inalizing he a icle. Ioanna G amma ikopoulou e iewed he p e ious li e a u e on ag icul u al landscape e alua ions. Timo Hu me desc ibed he linea mixed models and p o ided he analysis o Tables 3 and 4. Ka iina Soini pa icipa ed in planning he su ey and w i ing he In oduc ion and he Discussion and conclusions. Ma ja Uusi alo ook pa in planning he su ey and e iewed he li e a u e on landscape e alua ion me hods. Con lic o In e es The au ho s decla e no con lic s o in e es . Re e ences 1. Ruiz, J.; Domon, G. Rela ionships be ween u al inhabi an s and hei landscapes in a eas o in ensi e ag icul u al use: A case s udy in Quebec (Canada). J. Ru al S ud. 2012, 28, 590–602. 2. Soini, K.; Vaa ala, H.; Pou a, E. Residen s’ sense o place and landscape pe cep ions a he u al-u ban in e ace. Landsc. U ban Plan. 2012, 104, 124–134. 3. Kaljonen, M. Co-cons uc ion o agency and en i onmen al managemen . The case o ag i-en i onmen al policy implemen a ion a Finnish a ms. J. Ru al S ud. 2006, 22, 205–216. 4. Soini, K.; Aakkula, J. F aming he biodi e si y o ag icul u al landscape: The essence o local concep ions and cons uc ions. Land Use Policy 2007, 24, 311–321. 5. Council o Eu ope. The Eu opean Landscape Con en ion; Council o Eu ope: S asbou g, F ance, 2000. 6. Council o Eu ope. Fa o Con en ion on Value o Cul u al He i age on Socie y; Council o Eu ope: S asbou g, F ance, 2005. 7. Rosley, M.S.F.; Lami , H.; Rahman, S.R.A. Pe cei ing he aes he ic alue o he u al landscape h ough alid indica o s. P oced.—Soc. Beha . Sci. 2013, 85, 318–331. 8. F y, G.; T ei , M.S.; Ode, Å.; Vela de, M.D. The ecology o isual landscapes: Explo ing he concep ual common g ound o isual and ecological landscape indica o s. Ecol. Indic. 2009, 9, 933–947. 9. Van de Jag , A.; C aig, T.; Anable, J.; B ewe , M.J.; Pea son, D.G. Unea hing he pic u esque: The alidi y o he p e e ence ma ix as a measu e o landscape aes he ics. Landsc. U ban Plan. 2014, 124, 1–13. 10. T ei , M.S.; Ode, Å.; F y, G. Key concep s in a amewo k o analysing isual landscape cha ac e . Landsc. Res. 2006, 31, 229–256. 11. Se enan , M.; An op, M. Cogni i e a ibu es and aes he ic p e e ences in assessmen and di e en ia ion o landscapes. J. En i on. Manag. 2009, 90, 2889–2899. 12. Se enan . M.; An op, M. The use o la en classes o iden i y indi idual di e ences in he impo ance o landscape dimensions o aes he ic p e e ence. Land Use Policy 2010, 27, 827–842. 13. T ess, B.; T ess, G. Capi alizing on mul iplici y: A ansdisciplina y sys ems app oach o landscape esea ch. Landsc. U ban Plan. 2001, 57, 143–157. Land 2014, 3 614 14. Coe e ie , J.F. Dominan a ibu es in he pe cep ion and e alua ion o he Du ch landscape. Landsc. U ban Plan. 1996, 34, 27–44. 15. Sulli an, W.C.; Ande son, O.M.; Lo ell, S.T. Ag icul u al bu e s a he u al-u ban inge: An examina ion o app o al by a me s, esiden s, and academics in he Midwes e n Uni ed S a es. Landsc U ban Plan. 2004, 69, 299–313. 16. Tempes a, T. The pe cep ion o ag a ian his o ical landscapes: A s udy o he Vene o plain in I aly. Landsc. U ban Plan. 2010, 97, 258–272. 17. Bujis, A.E.; Ped oli, B.; Luginbühl, Y. F om hiking h ough a mland o a ming in a leisu e landscape: Changing social pe cep ions o he Eu opean landscape. Landsc. Ecol. 2006, 21, 375–389. 18. Rogge, E.; Ne ens, F.; Gulinck, H. Pe cep ion o u al landscapes in Flande s: Looking beyond aes he ics. Landsc. U ban Plan. 2007, 82, 159–174. 19. T ei , M.S. Indica o s o isual scale as p edic o s o landscape p e e ence; A compa ison be ween g oups. J. En i on. Manag. 2009, 90, 2882–2888. 20. Ode Sang, Å.; T ei , M.S. Pe cep ions o s ewa dship in No wegian ag icul u al landscapes. Land Use Policy 2013, 31, 557–564. 21. Sha e , J.E.L.; B ush, R.O. How o measu e p e e ences o pho og aphs o na u al landscapes. Landsc. Plan. 1977, 4, 237–256. 22. Daniel, T.C.; Vining, J. Me hodological issues in he assessmen o landscape quali y. In Beha iou and he Na u al En i onmen ; Al man, I., Wohwill, J.F., Eds.; Plenum P ess: New Yo k, NY, USA, 1983; pp. 39–83. 23. Sco , M.J.; Can e , D.V. Pic u e o place? A mul iple so ing s udy o landscape. J. En i on. Psychol. 1997, 17, 263–281. 24. Ka jalainen, E.; Komulainen, M. Field a o es a ion p e e ences: A case s udy in no heas e n Finland. Landsc. U ban Plan. 1998, 43, 79–90. 25. Daniel, T.C.; Mei ne , M.M. Rep esen a ional alidi y o landscape isualiza ions: The e ec s o g aphical ealism on pe cei ed scenic beau y o o es is as. J. En i on. Psychol. 2001, 21, 61–72. 26. Shu lewo h, S. The use o pho og aphs as an en i onmen al p esen a ion medium in landscape s udies. J. En i on. Manag. 1980, 11, 61–76. 27. Rose, G. Visual me hodologies. In An In oduc ion o he In e p e a ion o Visual Ma e ials, 2nd ed.; Sage Publica ions: London, UK, 2007; pp. 301. 28. Nassaue , J.I. F aming he landscape in pho og aphic simula ion. J En i on. Manag. 1983, 16, 4. 29. Mollie-S e ulesco, C.; Quesney, D. Séquences paysages. In Re ue de L’obse a oi e Pho og aphique du Paysage; Hazan, Éd.; Minis è e de l’en i onnemen : Pa is, F ance, 1997; p. 112. 30. T en , R.B.; Neumann, E.; K ashny, A. P esen a ion mode and ques ion o ma a i ac s in isual assessmen esea ch. Landsc. U ban Plan. 1987, 14, 225–235. 31. Ga cia Pé ez, J.D. Asce aining landscape pe cep ions and p e e ences wi h pai -wise pho og aphs: Planning u al ou ism in Ex emadu a, Spain. Landsc. Res. 2002, 27, 297–308. 32. He he ing on, J.; Daniel, T.C.; B own, T.C. Is mo ion mo e impo an han i sounds? The medium o p esen a ion in en i onmen al esea ch. J. En i on. Psychol. 1993, 13, 283–291. 33. Whe e , J.R. C ea ing landscape p e e ence models using in e ne su ey echniques. Landsc. Res. 2000, 25, 79–96. Land 2014, 3 615 34. Ro h, M. Valida ing he use o In e ne su ey echniques in isual landscape assessmen —An empi ical s udy om Ge many. Landsc U ban Plan. 2006, 78, 179–192. 35. Heikkilä, T.; Hie ala-Koi u, R. Maa alousmaiseman isuaalinen seu an a. In Maa alouden Ympä is ö uen Me ki ys Luonnon Monimuo oisuudelle ja Maisemalle: My as-Seu an a u kimus 2000–2003; In Se ies Suomen ympä is ö No. 709; Kuussaa i, M., Tiainen, J, Helenius, J., Hie ala- Koi u, R., Heliölä, J., Eds.; Edi a: Helsinki, Finland, 2004; p. 212. 36. Ode, Å.; T ei , M.S.; F y, G. Cap u ing landscapes isual cha ac e using indica o s: Touching base wi h landscape aes he ic heo y. Landsc. Res.2008, 33, 89–117. 37. Howley, P. Landscape aes he ics: Assessing he gene al publics’ p e e ences owa ds u al landscapes. Ecol. Econ. 2011, 72, 161–169. 38. Howley, P.; Hynes, S.; Donoghue, C.O. Coun yside p e e ences: Explo ing indi iduals’ willingness o pay o he conse a ion o he adi ional a m landscape. Landsc. Res. 2012, 37, 703–719. 39. D ams ad, W.E.; T ei , M.S.; Fjells ad, W.J.; F y, G.L.A. Rela ionships be ween isual landscape p e e ences and map-based indica o s o landscape s uc u e. Landsc. U ban Plan. 2006, 78, 465–474. 40. Benjamin, K.; Boucha d, A.; Domon, G. Abandoned a mlands as componen s o u al landscapes: An analysis o pe cep ions and ep esen a ions. Landsc. U ban Plan. 2007, 83, 228–244. 41. S umse, E.; Hauge, L. Landscape p o ec ion e alua ions and isual p e e ences o wes e n No wegian ag a ian landscapes. No sk Geog . Tidssk . 1998, 52, 1–15. 42. A iaza, M.; Cañas-O ega, J.F.; Cañas-Madueño, J.A.; Ruiz-A iles, P. Assessing he isual quali y o u al landscapes. Landsc. U ban Plan. 2004, 69, 115–125. 43. Rech man, O. Visual pe cep ion o ag icul u al cul i a ed landscapes: Key componen s as p edic o s o landscape p e e ences. Landsc. Res. 2013, 38, 273–294. 44. Sayadi, S.; Gonzalez, M.; Cala a a-Requena, J. Public p e e ences o landscape ea u es: The case o ag icul u al landscape in moun ainous Medi e anean a eas. Land Use Policy 2009, 26, 334–344. 45. Rambolinaza, M.; Dacha y-Be na d, J. Land-use planning and public p e e ences: Wha can we lea n om choice expe imen me hod? Landsc. U ban Plan. 2007, 83, 318–326. 46. G amma ikopoulou, I.; Pou a, E.; Salmio i a, M.; Soini, K. He e ogeneous p e e ences o ag icul u al landscape imp o emen s in sou he n Finland. J. Landsc. U ban Plan. 2012, 107, 181–191. 47. Campbell, D. Willingness o pay o u al landscape imp o emen s: Combining mixed logi and andom-e ec s models. J. Ag ic. Econ. 2007, 58, 467–483. 48. Van Be kel, D.B.; Ve bu g, P.H. Spa ial quan i ica ion and alua ion o cul u al ecosys em se ices in an ag icul u al landscape. Ecol. Indic. 2014, 37 (Pa A), 163–174. 49. Howley, P.; Donoghue, C.O.; Hynes, S. Explo ing public p e e ences o adi ional a ming landscapes. Landsc. U ban Plan. 2012, 104, 66–74. 50. Hynes, S.; Campbell, D. Es ima ing he wel a e o ag icul u al landscape change in I eland: A choice expe imen app oach. J. En i on. Plan. Manag. 2011, 54, 1019–1039. 51. A nbe ge , A.; Ede , R. Explo ing he he e ogenei y o u al landscape p e e ences: An image-based la en class app oach. Landsc. Res. 2011, 36, 19–40. Land 2014, 3 616 52. Snide , J.G.; Osgood, C.E. Seman ic Di e en ial Technique: A Sou cebook; Chicago: Aldine, TX, USA, 1969. 53. Múgica, M.; de Lucio, J.V. The ole o on-si e expe ience on landscape p e e ences. A case s udy a Doñana Na ional Pa k (Spain). J. En i on. Manag. 1996, 47, 229–239. 54. Kal enbo n, B.P.; Bje ke, T. Associa ions be ween en i onmen al alue o ien a ions and landscape p e e ences. Landsc. U ban Plan. 2002, 59, 1–11. 55. F anco, D.; F anco, D.; Mannino, I.; Zane o, G. The impac o ag o o es y ne wo ks on scenic beau y es ima ion: The ole o a landscape ecological ne wo k on a socio-cul u al p ocess. Landsc. U ban Plan. 2003, 62, 119–138. 56. I es, C.D.; Kendal, D. Values and a i udes o he u ban public owa ds pe i-u ban ag icul u al land. Land Use Policy 2013, 34, 80–90. 57. Ryan, R.L. Local pe cep ions and alues o a Midwes e n i e co ido . Landsc. U ban Plan. 1998, 546, 1–13. 58. C ow, T.; B own, T.; de Young, R. The Ri e side and Be wyn expe ience: Con as s in landscape s uc u e, pe cep ions o he u ban landscape, and hei e ec s on people. Landsc. U ban Plan. 2006, 75, 282–299. 59. Tah anainen, L.; Ty äinen, L.; Ihalainen, M.; Vuo ela, N.; Kolehmainen, O. Fo es managemen and public pe cep ions—Visual e sus e bal in o ma ion. Landsc. U ban Plan. 2001, 53, 53–70. 60. Tah anainen, L.; Ty äinen, L. Model o p edic ing he scenic alue o u al landscape: A p elimina y s udy o landscape p e e ences in No h Ca elia. Scand. J. Fo . Res. 1998, 13, 379–385. 61. Gómez-Limón, J.; de Lucío, J.V. Changes in use and landscape p e e ences on he ag icul u al-li es ock landscapes o he cen al Ibe ian Peninsula (Mad id, Spain). Landsc. U ban Plan. 1999, 44, 165–175. 62. B ush, R.; Chenowe h, R.E.; Ba man, T. G oup di e ences in he enjoyabili y o d i ing h ough u al landscapes. Landsc. U ban Plan. 2000, 47, 39–45. 63. Talous u kimus. In e ne Panel. A ailable online: h p://www. alous u kimus. i/in-english/ p oduc s_se ices/in e ne _panel/ (accessed on 28 No embe 2013). 64. S a is ics Finland. A ailable online: h p:// www.s a . i, 2011 (accessed on 28 No embe 2013). 65. Li ell, R.C.; Milliken, G.A.; S oup, W.W.; Wol inge , R.D.; Schabenbe ge , O. SAS® o Mixed Models, 2nd ed.; SAS Ins i u e Inc.: Ca y, NC, USA, 2006. 66. Bu on, R.J.F. Seeing h ough he “good a me ’s” eyes: Towa ds de eloping an unde s anding o he social symbolic alue o “p oduc i is ” beha iou . Sociol. Ru al. 2004, 44, 195–216. 67. Soini, K. Beyond he Ecological Ho Spo s: Unde s anding Local Residen s’ Pe cep ions o Biodi e si y o Ag icul u al Landscapes. Ph.D. Thesis, Tu un Yliopis o, Tu ku, Finland, 2007. © 2014 by he au ho s; licensee MDPI, Basel, Swi ze land. This a icle is an open access a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion license (h p://c ea i ecommons.o g/licenses/by/3.0/).