Preference for site conservation in relation to on-site biodiversity and perceived site attributes: An on-site survey of unmanaged urban greenery in a tropical city

Urban Forestry & Urban Greening 28 (2017) 12–20

Contents lists available at ScienceDirect

Urban Forestry & Urban Greening journal homepage: www.elsevier.com/locate/ufug

Original article

Preference for site conservation in relation to on-site biodiversity and perceived site attributes: An on-site survey of unmanaged urban greenery in a tropical city

MARK

⁎

Yun Hye Hwang , Charlotte J. Roscoe Department of Architecture, School of Design and Environment, National University of Singapore, 4 Architecture Drive, 117566, Singapore

A R T I C L E I N F O

A B S T R A C T

Keywords: Unmanaged urban greenery Biodiversity conservation On-site survey Landscape attributes Site conservation Singapore On-site biodiversity Perception study Tropical city Urban greenery Secondary forest Landscape preference

Singapore is a densely-populated tropical city with a heavy investment in managed greenery and intensive landscape maintenance practices. Unmanaged greenery such as unprotected young secondary forests continues to be lost in the face of continuous development despite its socio-ecological values. In this study, we assessed perceptions of 11 patches of unmanaged forested areas using an on-site questionnaire. The questionnaire covered site attributes pertaining to ecosystem services and disservices, as well as sensory experience, maintenance, and biodiversity, and recorded preference for retaining diverse forest patches in residential areas as opposed to replacing them with manicured greenery. To explore the association of on-site biodiversity with preference for site conservation, we surveyed flora and fauna at all 11 sites. We analysed the inter-relationship between perceived site attributes using Principal Component Analysis (PCA). The resulting factors, plus recorded site-level biodiversity, were then used as predictor variables in a confounder-adjusted binomial regression for preference for site conservation. The results highlight the joint importance of biodiversity, environmental function, and aesthetics in a stated preference for conservation of unmanaged urban greenery. We conclude with a discussion of design and management strategies aimed at balancing these site attributes and encouraging acceptance of unmanaged landscape in the tropical context.

1. Introduction Urbanization has caused the removal and extensive deterioration of urban forests in cities, even though an extensive body of literature notes their value. These include ecological, economic, socio-cultural, health, and environmental benefits (Jim and Chen, 2009; Konijnendijk, 2000; Roy et al., 2012; Mcdonnell and Kendal, 2015). Although the provision of urban ecosystem services is central to decision making in urban planning and policy processes (Gómez-Baggethun et al., 2013), and unmanaged urban secondary forests, are able to provide those services, they are often subject to competing interests between intensification of land development and environmental sustainability (Krajter Ostoić et al., 2015; Kowarik and Körner, 2005). Moreover, the factors that drive public support for conservation of urban forests are often overlooked. Many studies in the past four decades have addressed people’s preferences for nature (Kaplan et al., 1972; Ulrich, 1981; Kaplan and Kaplan, 1980; Berg et al., 2007; Han, 2010), with recent perception studies of urban forests citing a range of factors guiding preferences, from the size of greenery (Rink and Emmrich, 2005), to psychological ⁎

restoration (Hartig et al., 2006), health (Karjalainen et al., 2010), microclimate amelioration (Chen and Nakama, 2015), plant diversity (Mathey et al., 2016; Huang, 2014; Fuller et al., 2007; Lafortezza et al., 2008), quality of management (Ostoić et al., 2017), recreational value (Tyrväinen, 2001), landscape beauty (Huang, 2014), and economic benefits (Huang, 2014). On the other hand, urban forests are politically neglected (Gudurić et al., 2011), and detractors point to the misbehaviour of users (Ostoić et al., 2017), poor quality of maintenance (Verlič et al., 2015), or safety issues such as fear of crime (Jorgensen et al., 2005; Sreetheran and Van Den Bosch, 2014). In contrast, relatively few empirical studies about what drives people to conserve urban forests have been conducted. Such studies are even scarcer in the context of an Asian tropical city. To begin to address this deficit, this study used Singapore as a case study. Situated at the southern tip of the Malaysian peninsula (103°50′E, 1°20′N), Singapore is a tropical city-state with a total landmass of 719.1 km2. Over recent decades, large tracts of secondary rainforest have been removed with rapid urbanization and intensification (O'dempsey, 2014) and, as a result, natural terrestrial habitats have been subject to a “catastrophic rate of extinction” (Brook et al., 2003). Singapore’s widely-employed

Corresponding author. E-mail address: [email protected] (Y.H. Hwang).

http://dx.doi.org/10.1016/j.ufug.2017.09.011 Received 16 July 2017; Received in revised form 15 September 2017; Accepted 19 September 2017 Available online 21 September 2017 1618-8667/ © 2017 Elsevier GmbH. All rights reserved.

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

Fig. 1. Eleven sites are indicated in the map with site photos; the outermost North, East, South, and West sites are located respectively in Punggol, Bedok, Bukit Merah, and Bukit Batok planning areas.

Table rasa approach to development – combined with government-led strategic investment in heavily-managed and manicured greening (Neo et al., 2012) – continues to lead to the rapid transition of Singapore’s remaining unmanaged green spaces into curated gardens and lawns (Yee et al., 2011; Auger, 2013). Singapore provides an excellent case study for urban ecosystem services (Friess, 2016), and cultural ecosystem services such as recreation are of increasing importance in the country’s nature areas (Thiagarajah et al., 2015). Secondary forests that grew on abandoned agricultural land since the 1960s are typically dominated by exotic

species (Yee et al., 2016a), and are often overlooked in assessments of ecosystem services. However, several leading ecologists and environmentalists have highlighted the biodiversity conservation value of secondary forests in Singapore (Breugel et al., 2013; Tan et al., 2016b; NSS, 2011; Yee et al., 2016b; Koh and Sodhi, 2004). There has also been a steady growth in public concern, with petitions to protect secondary forests as socio-cultural assets (Tan et al., 2016b). The quantification of ecosystem services provided by greenery in a tropical context – such as the reduction of the urban heat island effect (Wong and Yu, 2005) and the amelioration of outdoor thermal comfort (Yang et al., 2013) – 13

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

Fig. 2. Interviewees’ characteristics (individual predisposition) were hypothesized to contribute to the interpretation of the physical site attributes and controlled for. This interpretation, in the form of perceived site attributes, was expected to predict preference for site conservation/retention of unmanaged greenery (preference for the site). Both the physical space and the perception can be targeted in design and management strategies to increase preferences for conservation.

species that are present is dependent on the seed source availability of the surroundings during the time of clearance, and succession” (Yee et al., 2016a). Eleven survey sites of this characterization were selected in the Central (6), East (2), West (1), and North East (2) planning regions of Singapore (see Fig. 1). Site selection was based on the following criteria; 1) the presence of a developed tree canopy with undergrowth in the form of forbs, grasses, and other groundcover; 2) the presence of low- to no-maintenance, unmanaged greenery, typically in the form of state-land that has been allowed to follow succession towards secondary forest; 3) a site in a residential location at a maximum proximity of 100 m to the nearest public housing areas, condominiums and/or landed housing estates; 4) an accessible site without enclosures or governmental trespassing restriction signs; and 5) land not classified as a park or nature reserve. Eleven survey sites meeting these criteria were selected in various geographical planning regions of Singapore. Eight of the 11 sites are earmarked for future development in the most recent Urban Redevelopment Authority (URA) Master Plan (URA, 2014).

support the retention of greenery in Singapore for human health and well-being. A survey administered by Khew et al. (2014) found Singaporean residents prefer low-diversity, manicured landscapes over high-diversity, naturalistic landscapes, but express seemingly contradictorily strong support for biodiversity conservation. Other assessments of residents’ perceptions of managed neighbourhood parks (Yuen, 1996), park connectors (Yuen et al., 1999), and green roofs (Yuen and Nyuk Hien, 2005) show that the public have some awareness of ecosystem services, as well as an appreciation of the beautification provided by greenery. The conflicted relationship between urbanites and natural or naturalistic vegetation is hypothesised to stem from a lack of familiarity with diverse and complex landscapes – particularly for younger generations in Singapore (Kong et al., 1999). Previous research has broached the topic of ecosystem disservices in Singapore and highlighted several perceived disservices of spontaneous vegetation on rooftops, including interviewee concerns for the facilitation of mosquito breeding, the potential harbouring of snakes and a general feeling of unease (Hwang and Roscoe, 2015). This study hypothesized that the perception of multiple site attributes contributes to residents’ preference for the retention of unmanaged greenery. It aimed to broadly investigate perceived site attributes in the tropical context, assess attribute inter-relationships and associations between perceived attributes and the preference to retain or remove vegetation. The objectives were: 1) to identify residents’ perceptions of unmanaged urban forest patches through on-site interviews; 2) to explore the inter-relationship of perceived site attributes; 3) to investigate if recorded biodiversity or perceived biodiversity are predictors of residents’ preference to retain or replace unmanaged greenery; and 4) to discuss ways to facilitate public acceptance of unmanaged, spontaneous vegetation, with a special consideration of biodiversity conservation and aesthetics. Results can be used to inform sustainable green space policies and management strategies that aim to integrate spontaneous, low- to no-maintenance greenery into the residential setting.

2.2. Questionnaire development A five-point Likert scale (from“ strongly agree” = 5 to “strongly disagree” = 1) was used to broadly cover ecosystem services and disservices identified in local perception studies (see previous research by Hwang and Roscoe, 2015; Yuen and Nyuk Hien, 2005), perception studies of aesthetics and classically-investigated landscape site attributes (e.g. Sevenant and Antrop, 2009; Kaplan and Kaplan, 1989; Tveit et al., 2006; Coeterier, 1996), and safety and crime concerns (e.g. Jorgensen et al., 2002; Yeoh and Yeow, 1997). Perceived biodiversity was measured in a similar manner to previous research by Fuller et al. (2007). In brief, four categories of species richness in a suitable range for plants, birds and butterflies were presented to interviewees. The questionnaire contained items covering demographics (age, gender, qualification level, and ethnicity) (Fig. 2). 2.3. On-site interviews

2. Methodology On-site interviews were conducted between February and June 2016, in fair weather only, between 4.30pm and 8pm to avoid the hottest hours of the day. Each interview lasted around 15 min and was conducted using the app Flocktracker, an in-field data collection platform developed by the MIT Alliance for Research and Technology. The survey was available in both an English and a Chinese language version (the latter was produced and verified with the help of Chinese-speaking staff at National University of Singapore). Every other passer-by was requested to complete the interview. The survey at each site ended

2.1. Survey setting Unmanaged greenery covers 28.5% of Singapore’s land, of which 19.6% is classified as young secondary forest (Yee et al., 2011). The sites under investigation in this study are classified as young secondary forest; more specifically, they are characterized as unmanaged wastewoodlands. These are “forests regrown on land that was usually cleared after the 1960s, and are dominated by exotic tree species, and whatever 14

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

“preference for conservation of the site,with retention of unmanaged greenery” – was carried out initially on only the scale items produced via PCA (treated as continuous) and the dichotomized (categorical) variable “messy” (“damp” showed inadequate correlation to the outcome variable and was therefore excluded). A second adjusted regression accounted for potential demographic covariates; age data were originally split into five-year intervals and treated as continuous; all other variables were treated as categorical, and education groups with insufficient representation (“other”, “no formal education/primary”) were excluded from analysis. Likelihood ratio testing of regression models against null models was done using the “lmtest” package (Zeileis and Hothorn, 2002), and multicolinearity was assessed using variance inflation factors and the “car” package (Fox and Weisberg, 2011).

Table 1 Participant characteristics(n = 327). Demographic variable

No.

%

Age 19 years and below 20–29 years 30–39 years 40–49 years 50–59 years 60 years and above Ethnicity Chinese Malay

51 95 65 42 43 31

15.6 29.1 19.9 12.9 13.1 9.4

199 30

60.8 9.2

Indian Other

52 46

15.9 14.1

Demographic variable Sex Female Male Highest education No formal education/Primary* Some secondary schooling Completed secondary school Completed JC/Pre-U/A-level Diploma/Professional University/Post-graduate degree Other*

No.

%

159 168

48.6 51.4

2 22 48 42 59 152

0.6 6.8 14.7 12.8 18.0 46.5

2

0.6

*Eliminated from regressions due to low representation.

3. Results

when around 30 people had completed the interview at each site (total n = 327; see Table 1 for demographic representation). Sample size was pre-determined by resource availability, and a relatively equal distribution of an adequate number of interviews across sites was aimed for in order to assess the effect of site on preference for conservation. The total sample was used in all final analyses. Interviews were conducted around 10 m from the edge of the vegetation across sites.

3.1. Participants A total of 327 on-site interviews were conducted between February and July 2016, with eight conducted in Chinese. There were no significant differences attributable to survey language, and overall response rate for the questionnaire was 67%. The majority of interviewees (68.2%) lived within 1 km of the site in question. Due to extremely low numbers of passers-by at the Punggol site (North), only eight interviews were conducted there. The range per site excluding Punggol was 30–37 interviewees. The average nature-relatedness score (5-point scale) of all interviewees was 3.81 (SD ± .65). *Eliminated from regressions due to low representation

2.4. On-site biodiversity survey Bird, butterfly, and plant species richness and bird and butterfly abundance were surveyed at each site between May and July 2016. Birds were surveyed at one hour after sunrise on two separate occasions per site. Butterfly survey transects were centered on the location of the avian point-count and ran along the edge of the site that was visible during interviews (40 m). Butterflies were actively searched for and recorded up to a 10 m width into the site from the 40 m transect line (Dallimer et al., 2012). Each survey lasted 15 min and was conducted between 9am and 10am (three surveys per site). Butterflies were netted and temporarily retained in clear plastic containers for identification purposes and to avoid double counting. Forbs (all trees, woody shrubs, herbaceous plants, climbers/creepers, ferns, palms, and epiphytes) in the 40 m × 10 m area were exhaustively surveyed and recorded at each site and identified using local keys and resources (Chen et al., 2015; Boo et al., 2014; Foo and Tan, 1986). All identification queries were resolved with the help of experts (see acknowledgements). Overhead tree cover was measured using a densitometer and averaged between three equally spaced points within the 40 m × 10 m area.

3.2. Recorded and perceived biodiversity In the recorded species richness data, bird, butterfly, and plant species richness measures were uncorrelated, however perceived richness measures were correlated for butterflies and birds (rs = .75, p = < .001) and plants and butterflies (rs = .73, p = .01). Recorded overhead tree cover (densitometer average) was negatively correlated with recorded butterfly species richness (rs =− .31, p = < .01) and recorded plant species richness (rs = − .57, p = .01) Recorded and perceived species richness was not significantly correlated across the three taxa, nor were recorded abundance and perceived species richness; however, perceived plant species richness was correlated with site-level densitometer readings for overhead tree cover (rs = .32, p = .03). Nature relatedness score (NR-6) of interviewee was positively correlated with perceived species richness (birds, rs = .67, p = < .001; butterflies, rs = .62, p = < .001; plants, rs = .53, p = < .001) (Table 2).

2.5. Analysis All analysis of data was carried out in RStudio. Spearman’s rank correlations were performed on the on-site recorded and perceived biodiversity data (see section 3.2.). The interrelationship of all perceived site attributes was explored using the “psych” package (Revelle, 2016). The number of factors to be retained in PCA analysis (see section 3.4.) was determined visually using a scree plot and based on Eigen values – all retained factors had an Eigen value above 2. Resulting factors were subsequently treated as scales for input into regression, and internal validity of the five scales was assessed using Cronbach's coefficient alpha. Items that double-loaded or did not load during PCA (“messy” and “damp”) were dichotomized for input into regressions. Note that “messy” was significant overall as a categorical variable in the final adjusted regression (Wald test for adjusted regression: Chisq. = 26.9, d.f. = 4, P = < .01), although it was dichotomized to avoid overcomplicating output by creating multiple inter-rank comparisons. Binomial linear regression for the outcome variable of interest –

Table 2 Biodiversity measures at sites. Recorded Biodiversity

Perceived Biodiversity

Average

Range

Average

Range

Plant species richness

43

32–53

Bird species richness

12

9–19

< 10 to > 300 < 4 to > 20

Butterfly species richness Bird abundance Butterfly abundance Overhead tree covera

8

3–21

44 7 79%

4–83 4–22 46.7% – 100%

Cat. 2 (10–100 types) Cat. 2 (5–14 types) Cat. 2 (4–10 types) – – –

< 5 to > 30 – – –

a Average densiometer reading for three point localities within biodiversity survey area.

15

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

Table 3 Interrelationship of perceived site attributes.

*

Table 4 Preference for site conservation.

Perceived site attribute

EV(%)

Factor 1: Disservices I am afraid of what is hiding inside this greenery I think this greenery increases the risk of being bitten by a snake I think this greenery increases the risk of dengue and other mosquito-borne diseases I think this greenery increases the risk of crime I associate this greenery with criminal activity I would feel unsafe at night near this greenery I think this greenery is dangerous I think this greenery harbors more mosquitoes than other types of greenery I think this greenery harbors more snakes than other types of greenery

31

Factor 2: Ecosystem services and aesthetics I think this greenery is valuable for conservation (of plants and animals) I think this greenery is valuable for environmental function and regulation I think this greenery is beautiful I think this greenery is valuable for the development and education of children I think this greenery is tranquil (peace and quiet) I think this greenery is natural This greenery is familiar to me

22

Factor 3: Mystery I think this greenery is mysterious I think this greenery is vast (covers a large area) I think this greenery is wild I think this greenery is dark I think this greenery is dense I think this greenery is varied (as oppose to homogenous and all the same)

18

Factor 4: Orderly I think this greenery is orderly (arranged) I think this greenery is open (I can see clearly through to the other side) I think this greenery is well maintained

15

Factor 5: Biodiversity Different types of butterfly Different types of bird Different types of plant

14

Loading

0.76 0.74

Preference item

Response

No.

%

Preference for conservation of site

Yes (retain unmanaged greenery) No (replace with managed greenery) No preferencea

264

80.7

46

14.1

17

5.2

0.70 a

0.69 0.69 0.67 0.68 0.66

Excluded from regressions.

managed greenery, such as a lawn or garden with a lawn (14.1%) (Table 4).

0.59

3.6. Association of perceived site attributes and preference for conservation of site

0.77

The use of the factors (see section 3.4) as scales was deemed adequately reliable for internal consistency (Cronbach's coefficient alpha (95% CIs): F1: .88, (.86–.90); F2: .78 (.75–.82); F3: .64 (.58–.70); F4: .65 (.58–.72); F5: .67 (.61–.73). Of all factors, only Factor 2 was retained as a predictor in the final binomial regression model for preference for conservation of the site (Tables 5 and 6). Interviewees who scored higher on this scale (which included items on biodiversity conservation, environmental functions, beauty, and educational value) had a higher likelihood of opting for conservation of the site (note: a large confidence interval indicated variability in this association). In contrast, interviewees who agreed the site was “messy” were less likely to opt for conservation and more likely to prefer that unmanaged greenery be replaced with highly-managed greenery such as a lawn or a garden with a lawn. Higher recorded plant species richness slightly reduced the likelihood of preference to conserve the site, whereas higher recorded butterfly and bird species richness slightly increased the likelihood. Demographic covariates were not significantly associated with preference for site conservation (Table 6). “Site” was not a significant predictor of preference for conservation, nor was there significant difference between sites for the outcome variable. “Site” was not included as a factor in the adjusted model because: 1) at some sites, no interviewees agreed the site was “messy”; and 2) there was collinearity with site-level biodiversity.

0.75 0.67 0.65 0.55 0.50 0.47 0.66 0.64 0.62 0.61 0.61 0.50

0.76 0.71 0.69 0.79 0.75 0.68

EV = Explained variance.

4. Discussion

3.3. Perceived site attributes

4.1. What drives people to appreciate unmanaged urban secondary forest? and how does this apply to conservation of unmanaged greenery?

Whilst observing unmanaged urban greenery, interviewees most agreed with the positive site attributes “natural”, “beautiful”, “valuable for environmental function and regulation”, and “valuable for conservation of plants and animals” (92%, 83%, 86%, 81%, respectively) and most disagreed with negative attributes (disservices), such as “criminal activities”, “crime risk”, and “danger” (9%, 8%, 11% respectively).

The items “high abundance of snakes” and “natural” were close to the cut-off for a second factor (“snakes” F5 = .42; “natural” F3 = .43). We hypothesize this is because snakes are not consistently viewed as a disservice; at times, they are seen as evidence of biodiversity; meanwhile, naturalness is not only associated with pleasing the beholder (F2) but can also be perceived as mysterious (F3). Over 80% of interviewees preferred to retain unmanaged greenery

3.4. Interrelationship of perceived site attributes

Table 5 Binomial regression of preference for conservation of the site, unadjusted for demographic variables.

Five site attribute factors were identified through factor analysis (Table 3): disservices (F1), ecosystem services and aesthetics (F2), mystery (F3), orderly (F4) and perceived biodiversity (F5). “I think this greenery is damp” did not load on to any factor. “I think this greenery is messy” loaded on two factors (negatively onto F2 and positively onto F3) and was therefore excluded from PCA. 3.5. Preference for conservation of the site The majority of interviewees wished to conserve the site’s unmanaged greenery (80.7%) rather than replace greenery with heavily-

Predictor variable

OR

95% CI

P-value

Factor 2 scale: Ecosystem services and aesthetics Messy: Agree* Plant species richness Butterfly species richness Bird species richness

3.24 0.26 0.92 1.09 1.15

1.74–6.26 0.12–0.55 0.86–0.98 1.02–1.17 1.02–1.31

< 0.01 < 0.01 0.01 0.02 0.02

*Reference category: Disagree; Likelihood ratio test for model: Chi-sq. = 44.50, d.f. = 5; P < .0001.

16

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

creation of “preferable” and “worthwhile to be protected” sites. In another example, Hwang et al. suggest the use of ecological design strategies of deforestation management in a tropical city context that are relevant to integration and conservation of unmanaged greenery in residential zones (2016).

Table 6 Binomial regression of preference for conservation of the site, adjusted for demographic variables. Predictor variable

OR

95% CI

P-value

Factor 2 scale: Ecosystem services and aesthetics Messy: Agree* Plant species richness Butterfly species richness Bird species richness

3.37 0.22 0.91 1.08 1.15

1.78–6.69 0.10–0.49 0.85–0.97 1.01–1.16 1.02–1.32

< 0.01 < 0.01 0.01 0.03 0.04

Age Gender: Male* Ethnicity: Indian* Ethnicity: Malay* Ethnicity: Other* Education: Some Secondary* Education: Completed secondary* Education: Polytechnic* Education: University*

0.97 0.82 1.57 1.00 0.76 0.43 0.45 1.51 1.06

0.87–1.09 0.39–1.70 0.56–4.99 0.30–4.10 0.30–2.16 0.09–1.91 0.11–1.79 0.37–6.15 0.30–3.30

0.63 0.60 0.41 1.00 0.59 0.27 0.26 0.56 0.92

4.2. What aspects of aesthetics are associated with unmanaged urban greenery? In our study, the site attributes that were rated the most positively pertained to naturalness, beauty, and ecosystem services. This is reminiscent of Meyer’s view of sustainable beauty as a channel for environmental appreciation and awareness (Meyer, 2008) or Jorgensen’s proposal of a future landscape aesthetics (Jorgensen, 2011) that connotes biological diversity as an aesthetic of necessity. Meanwhile, Gobster et al. point out that aesthetic pleasure has a complementary relationship with ecological health in landscape perception as long as it does not conflict with ecological goals or aesthetic preferences (2007) A majority of interviewees disagreed with site attributes “wildness” and “messiness” in our study, likely because the interview sites were at the periphery of unmanaged greenery; half of the scenery featured human-dominated landscapes with housing estates and neat lawns nearby as a cue to care. Similarly, if wildness in nature is to gain wider public acceptance, it needs to have recognizable ‘cues to care’ (Nassauer, 1995) and its ecological functions must be visualized. For example, information boards describing interventions may promote meaningful public-nature interactions and contribute to preference (Hill et al., 2007). If a balance can be struck between attractive and diversity-focused place-making near to where citizens live and work, support for species conservation may be engendered and human wellbeing enhanced (Miller, 2005). To further promote acceptance of unmanaged secondary forest near residential areas, designers could integrate colourful and well-maintained landscapes as signs of human intervention (Jorgensen et al., 2007) and introduce wildflowers at the periphery to increase aesthetic value (Kühn, 2006). We suggest future landscape research should look beyond comparisons with heavily-managed greenery and use experimental approaches to increase biodiversity (e.g. Shwartz et al., 2014) and alter edge aesthetics to increase perceptions of human intervention and care.

*Reference categories: Messy: Disagree; Gender: Female; Ethnicity: Chinese; Education: Junior College (equivalent qualification to Baccalaureate/A-levels); Likelihood ratio test for model: Chi-sq. = 51.16, d.f. = 14; P < .0001.

on the site rather than replace it with managed greenery (e.g. a lawn or garden with a lawn). This is inconsistent with previous studies of preferred forms of urban nature which concluded that urban dwellers prefer human-modified urban nature over wild nature (Lafortezza et al., 2008; Hofmann et al., 2012) and lean towards an aesthetic preference for neat, low-diversity greenery (Khew et al., 2014). A plausible explanation for this may be a familiarity with, and potential attachment to, the site in its present state (stasis), a concept known to be linked to landscape preference (Tang et al., 2014; Kaplan and Kaplan, 1989; Purcell et al., 2001). Our findings are congruent with a recent review of unmanaged urban greenery by Tan et al. (2016a). These researchers found the socio-cultural values attached to the forested landscape in Singapore have driven a surge in support for its protection. As shown in the high rating of ecosystem services by a majority of survey participants, public valuation extends beyond cultural values to more tangible ecosystem services. This is consistent with findings in other studies of willingness to conserve sites, namely that pragmatic needs are preferred over ethical dimensions (Lo and Jim, 2010), and people care only if there is direct relevance to their daily lives (Miller, 2005). The highly rated experiential and functional items of Factor 2 that, combined, predict preference for conservation of the site (regression analysis), indicate a good level of public awareness of the ecological and environmental functions of contact with nature. The link between knowledge of the ecological and environmental benefits of the forested landscape and preference for conservation, with the acceptance of wilder landscapes to achieve these ends, is documented in other research (e.g. Weber et al., 2014) and echoed in our findings. Nassauer and Raskin (2014) suggest that new design and planning approaches should be informed by an urban ecological knowledge that is synthesized with a social and cultural understanding of residents’ perceptions and values. Such an ecological knowledge includes Forman’s urban ecology principles (Forman, 2014), sustainability principles (Cadenasso and Pickett, 2008), ecology in design and planning (Melnick, 2001), and ecological design strategies (Spirn, 1985). To increase awareness and encourage acceptance of unmanaged greenery in the residential landscape up to a city scale, ecological and environmental functions should be considered in its designs and plans – this is crucial to encourage a positive responses from the public. For example, Li et al.’s paper (2005) on comprehensive concept planning of urban greening is based on ecological principles that consider urban greenery to be part of a city’s ecological network of urban biodiversity conservation at a nested scale (Li et al., 2005). Using urban greenery to perform biophysical functions, such as urban heat mitigation, flood control, and storm water management, would work towards the

4.3. To what extent does the perceived and recorded biodiversity affect responses to site conservation? In this study, there was no clear contradiction between biodiversity support and preference for low-diversity, high-maintenance greenery, as suggested in a survey by Khew et al. (2014). However, preference for conservation of the unmanaged greenery was not predicted specifically by perceived species richness of the three taxa (plants, birds, butterflies). This suggests either that the interviewees were not adept at accurately perceiving biodiversity (Pett et al., 2016; Shwartz et al., 2014) or that biodiversity for biodiversity’s sake (i.e. with no perceived benefit to residents) was not relevant to their stated preferences i.e. biodiversity may be perceived in terms of wider ecosystem services and aesthetics and irrelevant at the scale of individual taxa. Dallimer et al. (2012) found a different relationship when examining interviewee wellbeing; in their study, perceived diversity rather than recorded biodiversity was significantly associated with wellbeing. We hypothesize, in agreement with previously referenced studies, that this is due to large objects, overall composition, and tree cover being used as a proxy for plant diversity by observers. In reality, tree cover and plant diversity show an inverse relationship because a dense tree cover limits light penetration and results in a lesser-developed, and varied understory that, in turn, lowers butterfly diversity. This inconsistent relationship between perceived and recorded biodiversity has been coined the “people-biodiversity paradox” (Pett et al., 17

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

ecosystem services and preferences (Bixler and Floyd, 1997) and may allow the identification of target groups for education and participatory activities. For example, people sometimes have contradictory feelings of appreciation and apprehension; they may value a wild landscape but are fearful of such things as mosquito-borne diseases or personal safety. Designers could reduce negative feelings by providing safe footpaths and creating design strategies to regulate mosquito populations, for example, avoiding stagnant water in the landscape (Knight et al., 2003), providing habitats for species that prey on mosquitoes, such as fish, insects, birds, and bats (Murdoch et al., 1985; Kunz et al., 2011).

2016) and has serious implications for aligning recorded biodiversity conservation and human well-being agendas. The benefit of light-based intervention by landscape designers to increase biodiversity in edge habitats should be investigated further in experiment-based research. Further investigation of the role of people’s connection to nature and other potential predispositions towards conservation is warranted given the strong positive correlation between nature-relatedness and perceived biodiversity. We interpret this as an indication that assessment of species richness depends not only on physical site attributes but also on what people expect to see in that particular setting based on their knowledge (Qiu et al., 2013), previous experiences (Conway and Yip, 2016), and relationship to nature (Hofmann et al., 2012). Previously, researchers have asked participants to self-define as “nature orientated” (Gunnarsson et al., 2017; Weber et al., 2014); we recommend the use of validated psychological scales, such as the short form NR-6 to improve future comparability. We found that preference for site conservation is positively related with fauna diversity over floral diversity. The goal of intervention should be clear: certain interventions may benefit site acceptability and increase fauna biodiversity, for example, planting native trees and shrubs for pollinators, planting fruit trees for a food source, adding bird habitats (bird houses, perches), and introducing attractive butterfly species and supporting flora would help increase diversity (and potentially increase preference for site conservation).

4.5. Limitations and research recommendations There are several limitations to our study. First, we attempted to cover a wide range of site attributes, and this necessarily limited the depth of investigation. We did not survey the entire forested area, and we were limited to one type of unmanaged greenery. Second, the focus of investigation was on perceived site attributes; excluding biodiversity measures, we did not quantify other potentially relevant on-site variables. Third, in the deployment of the questionnaire, there may have been biases in the responses; interviewees may have responded with answers they thought the interviewer wished to hear; it is also possible that potential interviewees with negative perceptions of the site were not interviewed as they selected a different walking route to avoid passing the site edge. Fourth, preference for conservation of a site with retention of unmanaged greenery may be explained by preference for stasis over change, not preference for a greenery management strategy. The specific study setting offers new insights into perceptions of unmanaged vegetation in the Asian tropics, though entails limited comparability with temperate settings where the term ‘unmanaged greenery’ often refers to herbaceous cover. Secondary forest grows rapidly in the Asian tropics and we propose further investigation of perceptions of different forest types, both nationally and internationally. We also have a number of recommendations for future work on unmanaged greenery. First, further research into unmanaged greenery perceptions using stratified samples would allow specific demographic groups to be identified and targeted for management strategies or awareness campaigns, although our study finds no strong evidence of an association between demographic variables and preference for conservation. Second, our biodiversity surveys covered only three taxa,

4.4. What site attributes are the least appreciated and to what extent can disservices be resolved? Several studies have recently argued for the importance of examining the negative side of ecosystem services to ensure a comprehensive assessment for decision making and urban greenery management (Delshammar et al., 2015; Lyytimäki et al., 2008) In our results, disservices loaded on a single factor (Table 3), and a strong majority of participants disagreed with the disservice statements (Fig. 3). Irrespective of the actual associated risk, perception of ecosystem disservices may influence public preferences (see review by Von Döhren and Haase, 2015). Therefore, recording these perceptions and working with the minority of concerned residents is essential to provide an inclusive landscape that can be enjoyed by all (Conway and Yip, 2016). The investigation of disservices complements research into

Fig. 3. Perceived site attributes across all 11 sites (ordered by “strongly agree” responses).

18

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

Cadenasso, M.L., Pickett, S.T., 2008. Urban principles for ecological landscape design and maintenance: scientific fundamentals. Cities Environ. (CATE) 1, 4. Chen, B., Nakama, Y., 2015. Residents’ preference and willingness to conserve homestead woodlands: coastal villages in Okinawa Prefecture, Japan. Urban For. Urban Greening 14, 919–931. Chen, L., Foong, A.W., NG, A., Teo, J., Tang, J., 2015. 1001 Garden Plants In Singapore. NParks, Singapore. Coeterier, J.F., 1996. Dominant attributes in the perception and evaluation of the Dutch landscape. Landscape Urban Plann. 34, 27–44. Conway, T.M., Yip, V., 2016. Assessing residents’ reactions to urban forest disservices: a case study of a major storm event. Landscape Urban Plann. 153, 1–10. Dallimer, M., Irvine, K.N., Skinner, A.M.J., Davies, Z.G., Rouquette, J.R., Maltby, L.L., Warren, P.H., Armsworth, P.R., Gaston, K.J., 2012. Biodiversity and the feel-good factor: understanding associations between self-reported human well-being and species richness. Bioscience 62, 47–55. Delshammar, T., Östberg, J., Öxell, C., 2015. Urban trees and ecosystem disservices-A pilot study using complaints records from three swedish cities. Arboric. Urban For. 41. Foo, T.S., Tan, B.H., 1986. A Guide to the Wildflowers of Singapore. Singapore Science Centre, Singapore. Forman, R.T., 2014. Urban Ecology: Science of Cities. University Press, Cambridge. Fox, J., Weisberg, S., 2011. An R Companion to Applied Regression. Sage, Thousand Oaks, CA. Friess, D.A., 2016. Singapore as a long-term case study for tropical urban ecosystem services. Urban Ecosyst. Fuller, R.A., Irvine, K.N., Devine-Wright, P., Warren, P.H., Gaston, K.J., 2007. Psychological benefits of greenspace increase with biodiversity. Biol. Lett. 3, 390–394. Gómez-Baggethun, E., Gren, Å., Barton, D.N., Langemeyer, J., mcphearson, T., O’farrell, P., Andersson, E., Hamstead, Z., Kremer, P., 2013. Urban ecosystem services. In: Elmqvist, T., Fragkias, M., Goodness, J., Güneralp, B., Marcotullio, J.P., Mcdonald, I.R., Parnell, S., Schewenius, M., Sendstad, M., Seto, C.K., Wilkinson, C. (Eds.), Urbanization, Biodiversity and Ecosystem Services: Challenges and Opportunities: A Global Assessment. Springer Netherlands, Dordrecht. Gobster, P.H., Nassauer, J.I., Daniel, T.C., Fry, G., 2007. The shared landscape: what does aesthetics have to do with ecology? Landscape Ecol. 22, 959–972. Gudurić, I., Tomićević, J., Konijnendijk, C.C., 2011. A comparative perspective of urban forestry in Belgrade, Serbia and Freiburg, Germany. Urban For. Urban Greening 10, 335–342. Gunnarsson, B., Knez, I., Hedblom, M., Sang, Å.O., 2017. Effects of biodiversity and environment-related attitude on perception of urban green space. Urban Ecosyst. 20, 37–49. Han, K.-T., 2010. An exploration of relationships among the responses to natural scenes: scenic beauty, preference, and restoration. Environ. Behav. 42, 243–270. Hartig, T., Staats, H., Institutet För Bostads-och, U., Humanistisk-Samhällsvetenskapliga, V., Samhällsvetenskapliga, F., Uppsala, U., 2006. The need for psychological restoration as a determinant of environmental preferences. J. Environ. Psychol. 26, 215–226. Hill, J., Woodland, W., Gough, G., 2007. Can visitor satisfaction and knowledge about tropical rainforests be enhanced through biodiversity interpretation, and does this promote a positive attitude towards ecosystem conservation? J. Ecotour. 6, 75–85. Hofmann, M., Westermann, J.R., Kowarik, I., Van Der Meer, E., 2012. Perceptions of parks and urban derelict land by landscape planners and residents. Urban For. Urban Greening 11, 303–312. Huang, S.-C.L., 2014. Park user preferences for establishing a sustainable forest park in Taipei, Taiwan. Urban For. Urban Greening 13, 839–845. Hwang, Y.H., Roscoe, C., 2015. Session 1R: Evaluating Native and Wild green Roof Performance and Desirability—Perceptions of a Wild Green Roof in Singapore. CitiesAlive, New York October 5–6. Hwang, Y.H., Feng, Y., Tan, P.Y., 2016. Managing deforestation in a tropical compact city (Part B) urban ecological approaches to landscape design. Smart Sustainable Built Environ. 5, 73–92. Jim, C.Y., Chen, W.Y., 2009. Ecosystem services and valuation of urban forests in China. Cities 26, 187–194. Jorgensen, A., Hitchmough, J., Calvert, T., 2002. Woodland spaces and edges: their impact on perception of safety and preference. Landscape Urban Plann. 60, 135–150. Jorgensen, A., Hitchmough, J., Dunnett, N., 2005. Living in the urban wildwoods: a case study of Birchwood, Warrington New Town, UK. Wild Urban Woodlands. Springer. Jorgensen, A., Hitchmough, J., Dunnett, N., 2007. Woodland as a setting for housingappreciation and fear and the contribution to residential satisfaction and place identity in Warrington New Town, UK. Landscape Urban Plann. 79, 273–287. Jorgensen, A., 2011. Beyond the view: future directions in landscape aesthetics research. Landscape Urban Plann. 100, 353–355. Kühn, N., 2006. Intentions for the unintentional: spontaneous vegetation as the basis for innovative planting design in urban areas. J. Landscape Archit. 1, 46–53. Kaplan, S., Kaplan, R., 1980. Cognition and Environment: Functioning in an Uncertain World. Ulrich's Bookstore, Ann Arbor, Mich. Kaplan, R., Kaplan, S., 1989. The Experience of Nature: A Psychological Perspective. University Press, Cambridge. Kaplan, S., Kaplan, R., Wendt, J.S., 1972. Rated preference and complexity for natural and urban visual material. Percept. Psychophys. 12, 354–356. Karjalainen, E., Sarjala, T., Raitio, H., 2010. Promoting human health through forests: overview and major challenges. Environ. Health Preventative Med. 15, 1–8. Khew, J.Y.T., Yokohari, M., Tanaka, T., 2014. Public perceptions of nature and landscape preference in Singapore. Hum. Ecol. 42, 979–988. Knight, R.L., Walton, W.E., O’meara, G.F., Reisen, W.K., Wass, R., 2003. Strategies for

but this could be extended to other perceptible fauna such as bees (Gunnarsson et al., 2017). Third, a comparison of heavily-managed greenery and unmanaged greenery in the same locale would be interesting; we opted to focus on unmanaged greenery partly because of our limited resources and partly because it is under-represented in research in Singapore; we were also concerned with the planned extensive development of unmanaged green spaces (Tan et al., 2016a). Finally, interview location at a site could be varied. For example, varying the distance from the interview location and the proportion of the forest and lawns visible from the interview location at a site may influence survey responses. 5. Conclusion Research on perceptions of biodiversity and landscape is often conducted in temperate regions with limited urban growth (Botzat et al., 2016). To the best of our knowledge, this is the first on-site perception study focusing on unmanaged urban secondary forest in Singapore. Due to its high biodiversity, small area, tropical location, and recent and continuing urbanization, Singapore offers an excellent case study for research into the link between biodiversity, health, and wellbeing (Sandifer et al., 2015). As public participation in planning and governance is set to increase in the country in the coming years (Noh and Tumin, 2008; Soh and Yuen, 2006), the assessment of sociocultural values and public perceptions of ecosystem services and disservices of unmanaged urban green spaces will be integral in designing landscape management strategies for the public benefit. Understanding public preferences and perceptions is vital in making landscape management and conservation efforts participatory in Singapore. Unmanaged urban greenery is an important part of urban ecosystems and secondary forest provides key ecosystem services. If the general public were to accept the conservation value of this type of greenery over lawns, it would increase biodiversity, foster the direct experience of nature by urban residents, and increase resident satisfaction in a tropical biophilic city. Acknowledgements This work was made possible by funding from NUS-MOE (Ministry of Education) Tier 1 Research Fund under Grant Number R-295-000112-112. Thank you to Chloe Tan (NUS) for her work on the biodiversity surveys and offering her ornithological expertise to the project. Thanks also to Anuj Jain (NUS) for his help with butterfly identification and Dr. Chong Kwek Yan (NUS), Professor Hugh Tan (NUS), Alex Yee (NParks), Ang Wee Foong (NParks) and Dr. Adrian Loo (NParks) for contributing their plant identification expertise. A special thanks to Eleanor Cowell and Poh QiYing (NUS) for conducting a large number of the on-site interviews, and to Jonathan Yue (NUS) for his advice and reviews of the paper. Thank you, finally, to Daniel Heriberto Palencia Arreola for providing technical support with the Flocktracker app. References Auger, T., 2013. Living in a Garden: The Greening of Singapore. Editions Didier Millet. Berg, V.D.A.E., Hartig, T., Staats, H., 2007. Preference for nature in urbanized societies: stress, restoration, and the pursuit of sustainability. J. Soc. Issues 63, 79–96. Bixler, R.D., Floyd, M.F., 1997. Nature is scary, disgusting, and uncomfortable. Environ. Behav. 29, 443–467. Boo, C.M., Chew, S.Y.J., Yong, J.W.H., 2014. Plants In Tropical Cities. Nature's Niche, Singapore. Botzat, A., Fischer, L.K., Kowarik, I., 2016. Unexploited opportunities in understanding liveable and biodiverse cities. A review on urban biodiversity perception and valuation. Global Environ. Change 39, 220–233. Breugel, M.V., Hall, J.S., Craven, D., Bailon, M., Hernandez, A., Abbene, M., Breugel, P.V., 2013. Succession ephemeral secondary forests and their limited role for the conservation of floristic diversity in a human-modified tropical landscape. Plos One 8, e82433. Brook, B.W., Sodhi, N.S., NG, P.K.L., 2003. Catastrophic extinctions follow deforestation in Singapore. Nature 424, 420–426.

19

Urban Forestry & Urban Greening 28 (2017) 12–20

Y.H. Hwang, C.J. Roscoe

Research. Northwestern University Evanston, Illinois R package version 1.6.12. Rink, D., Emmrich, R., 2005. Surrogate nature or wilderness? social perceptions and notions of nature in an urban context. Wild Urban Woodlands. Springer. Roy, S., Byrne, J., Pickering, C., 2012. A systematic quantitative review of urban tree benefits, costs, and assessment methods across cities in different climatic zones. Urban For. Urban Greening 11, 351–363. Sandifer, P.A., Sutton-Grier, A.E., Ward, B.P., 2015. Exploring connections among nature, biodiversity, ecosystem services, and human health and well-being: opportunities to enhance health and biodiversity conservation. Ecosyst. Serv. 12, 1–15. Sevenant, M., Antrop, M., 2009. Cognitive attributes and aesthetic preferences in assessment and differentiation of landscapes. J. Environ. Manage. 90, 2889–2899. Shwartz, A., Turbé, A., Simon, L., Julliard, R., 2014. Enhancing urban biodiversity and its influence on city-dwellers: an experiment. Biol. Conserv. 171, 82–90. Soh, E.Y., Yuen, B., 2006. Government-aided participation in planning Singapore. Cities 23, 30–43. Spirn, A.W., 1985. The granite garden: urban nature and human design. Basic Books. Sreetheran, M., Van Den Bosch, C.C.K., 2014. A socio-ecological exploration of fear of crime in urban green spaces-a systematic review. Urban For. Urban Greening 13, 1–18. Tan, P.Y., Feng, Y., Hwang, Y.H., 2016a. Deforestation in a tropical compact city (Part A). Smart Sustainable Built Environ. 5, 47–72. Tan, P.Y., Feng, Y., Hwang, Y.H., 2016b. Deforestation in a tropical compact city (part A). Smart Sustainable Built Environ. 5, 47–72. Tang, I.-C., Sullivan, W.C., Chang, C.-Y., 2014. Perceptual evaluation of natural landscapes: the role of the individual connection to nature. Environ. Behav. Thiagarajah, J., Wong, S.K.M., Richards, D.R., Friess, D.A., 2015. Historical and contemporary cultural ecosystem service values in the rapidly urbanizing city state of Singapore. Ambio 44, 666–677. Tveit, M., Ode, Å., FRY, G., 2006. Key concepts in a framework for analysing visual landscape character. Landscape Res. 31, 229–255. Tyrväinen, L., 2001. Economic valuation of urban forest benefits in Finland. J. Environ. Manage. 62, 75–92. URA, 2014. Urban Redevelopment Authority Master Plan 2014. Accessed 16 January 2015. www.ura.gov.sg/uol/ master-plan/view-master-plan/master-plan-2014/ master-plan.aspx. Ulrich, R.S., 1981. Natural versus urban scenes: some psychophysiological effects. Environ. Behav. 13, 523–556. Verlič, A., Arnberger, A., Japelj, A., Simončič, P., Pirnat, J., 2015. Perceptions of recreational trail impacts on an urban forest walk: a controlled field experiment. Urban For. Urban Greening 14, 89–98. Von Döhren, P., Haase, D., 2015. Ecosystem disservices research: a review of the state of the art with a focus on cities. Ecol. Indic. 52, 490–497. Weber, F., Kowarik, I., Säumel, I., 2014. A walk on the wild side: perceptions of roadside vegetation beyond trees. Urban For. Urban Greening 13, 205–212. Wong, N.H., Yu, C., 2005. Study of green areas and urban heat island in a tropical city. Habitat Int. 29, 547–558. Yang, W., Wong, N.H., Jusuf, S.K., 2013. Thermal comfort in outdoor urban spaces in Singapore. Build. Environ. 59, 426–435. Yee, A.T.K., Corlett, R.T., Liew, S.C., Tan, H.T.W., 2011. The vegetation of Singapore—an updated map. Gard. Bull. Singapore 63, 205–212. Yee, A.T.K., Chong, K.Y., Neo, L., Tan, H.T.W., 2016a. Updating the classification system for the secondary forests of Singapore. Raffles Bull. Zool. 11–21. Yee, A.T.K., Chong, K.Y., Neo, L., Tan, H.T.W., 2016b. Updating the classification system for the secondary forests of Singapore. Raffles Bull. Zool. 11–21. Yeoh, B.S.A., Yeow, P.L., 1997. Where women fear to tread: images of danger and the effects of fear of crime in Singapore. GeoJournal 43, 273–286. Yuen, B., Nyuk Hien, W., 2005. Resident perceptions and expectations of rooftop gardens in Singapore. Landsc. Urban Plann. 73, 263–276. Yuen, B., Kong, L., Briffett, C., 1999. Nature and the Singapore resident. GeoJournal 49, 323–331. Yuen, B., 1996. Use and experience of neighborhood parks in Singapore. J. Leisure Res. 28, 293–311. Zeileis, A., Hothorn, T., 2002. Diagnostic checking in regression relationships. R News 2, 7–10.

effective mosquito control in constructed treatment wetlands. Ecol. Eng. 21, 211–232. Koh, L.P., Sodhi, N.S., 2004. Importance of reserves, fragments, and parks for butterfly conservation in a tropical urban landscape. Ecol. Appl. 14, 1695–1708. Kong, L., Yuen, B., Sodhi, N.S., Briffett, C., 1999. The construction and experience of nature: perspectives of urban youths. Tijdschrift voor economische en sociale geografie 90, 3–16. Konijnendijk, C.C., 2000. Adapting forestry to urban demands—role of communication in urban forestry in Europe. Landscape Urban Plann. 52, 89–100. Kowarik, I., Körner, S., 2005. Wild Urban Woodlands: New Perspectives for Urban Forestry. Springer, New York; Berlin. Krajter Ostoić, S., Konijnendijk van den Bosch, C.C., Sveriges, L., 2015. Exploring global scientific discourses on urban forestry. Urban For. Urban Greening 14, 129–138. Kunz, T.H., Braun De Torrez, E., Bauer, D., Lobova, T., Fleming, T.H., 2011. Ecosystem services provided by bats. Ann. N. Y. Acad. Sci. 1223, 1–38. Lafortezza, R., Corry, R.C., Sanesi, G., Brown, R.D., 2008. Visual preference and ecological assessments for designed alternative brownfield rehabilitations. J. Environ. Manage. 89, 257–269. Li, F., Wang, R., Paulussen, J., Liu, X., 2005. Comprehensive concept planning of urban greening based on ecological principles: a case study in Beijing, China. Landscape Urban Plann. 72, 325–336. Lo, A.Y., Jim, C., 2010. Willingness of residents to pay and motives for conservation of urban green spaces in the compact city of Hong Kong. Urban For. Urban Greening 9, 113–120. Lyytimäki, J., Petersen, L.K., Normander, B., Bezák, P., 2008. Nature as a nuisance?: ecosystem services and disservices to urban lifestyle. Environ. Sci. 5, 161–172. Mathey, J., Arndt, T., Banse, J., Rink, D., 2016. Public perception of spontaneous vegetation on brownfields in urban areas—results from surveys in Dresden and Leipzig (Germany). Urban For. Urban Greening. Mcdonnell, M.J., Kendal, D., 2015. The ecology of urban forests. In: PEH, K.S.H., Corlett, R., Bergeron, Y. (Eds.), Routledge Handbook of Forest Ecology. Routledge, New York. Melnick, R., 2001. Ecology and Design: Frameworks for Learning. Island Press. Meyer, E.K., 2008. Sustaining beauty. The performance of appearance: a manifesto in three parts. J. Landscape Archit. 3, 6–23. Miller, J.R., 2005. Biodiversity conservation and the extinction of experience. Trends Ecol. Evol. 20, 430–434. Murdoch, W.W., Chesson, J., Chesson, P.L., 1985. Biological control in theory and practice. Am. Nat. 125, 344–366. Nss, 2011. Nature Society (Singapore)’s Position on Bukit Brown. https://www.nss.org. sg/documents/Nature%20Society's%20Position%20on%20Bukit%20Brown.pdf. Nassauer, J.I., Raskin, J., 2014. Urban vacancy and land use legacies: a frontier for urban ecological research, design, and planning. Landscape Urban Plann. 125, 245–253. Nassauer, J.I., 1995. Messy ecosystems, orderly frames. Landscape J. 14, 161–170. Neo, B.S., Gwee, J., Mak, C., 2012. Growing a city in a garden. In: GWEE, J. (Ed.), Case Studies in Public Governance. Building Institutions in Singapore. Routledge, Oxon. Noh, A., Tumin, M., 2008. Remaking public participation: the case of Singapore. In: In: Noh, A., Tumin, M. (Eds.), Remaking Public Participation: The Case of Singapore. Asian Social Science 4. pp. 19. O'dempsey, T., 2014. Singapore's changing landscape since c. 1800. In: O'dempsey, T., Emmanuel, M., Van Wyhe, J., Taylor, N.P., Tan, F.L., Chou, C., Yi, G.H., Heng, C. (Eds.), Nature Contained: Environmental Histories of Singapore. NUS Press, Singapore. Ostoić, S.K., Van Den Bosch, C.C.K., Vuletić, D., Stevanov, M., Živojinović, I., MutabdžijaBećirović, S., Lazarević, J., Stojanova, B., Blagojević, D., Stojanovska, M., 2017. Citizens’ perception of and satisfaction with urban forests and green space: results from selected Southeast European cities. Urban For. Urban Greening 23, 93–103. Pett, T.J., Shwartz, A., Irvine, K.N., Dallimer, M., Davies, Z.G., 2016. Unpacking the people-biodiversity paradox: a conceptual framework. Bioscience 66, 576–583. Purcell, T., Peron, E., Berto, R., 2001. Why do preferences differ between scene types? Environ. Behav. 33, 93–106. Qiu, L., Lindberg, S., Nielsen, A.B., 2013. Is biodiversity attractive?: —on-site perception of recreational and biodiversity values in urban green space. Landscape Urban Plann. 119, 136–146. Revelle, W., 2016. Psych: Procedures for Psychological, Psychometric, and Personality

20