Vegetation communities on commercial developments are heterogenous and determined by development and landscaping decisions, not socioeconomics


Authors: Karen Dyson aff001
Authors place of work: Research and Design for Integrated Ecology, Seattle, Washington, United States of America aff001;  Urban Ecology Research Laboratory, University of Washington, Seattle, Washington, United States of America aff002
Published in the journal: PLoS ONE 14(9)
Category: Research Article
doi: 10.1371/journal.pone.0222069

Summary

In urban ecosystems, woody vegetation communities and the ecosystem functions and habitat they provide are largely controlled by humans. These communities are assembled during development, landscaping, and maintenance processes according to decisions made by human actors. While vegetation communities on residential land uses are increasingly well studied, these efforts generally have not extended to other land uses, including commercial property. To fill this gap, I surveyed tree and shrub communities on office developments located in Redmond and Bellevue, Washington, USA, and explored whether aggregated neighborhood and parcel scale socio-economic variables or variables describing the outcome of development and landscaping actions better explained variation in vegetation communities. I found that both tree and shrub communities on office developments are heterogenous, with sites characterized by native or ornamental vegetation. The heterogeneity I observed in vegetation communities within one land use suggests that different ecosystem functions, habitat quality, and habitat quantities are provided on office developments. Greater provision of e.g. native conifer habitat is possible using currently existing developments as models. Additionally, the outcome of development and landscaping decisions explained more variation in community composition than the socio-economic factors found significant on residential property. Together with previous research showing that residential property owner attitudes and actions are more important than socio-economic descriptors, my results suggest that individual motivators, including intended audience, may be the primary determinant of urban vegetation communities. Future urban ecology research should consider sampling the vegetation gradient within land uses, better understanding individual motivation for vegetation management, and creating models of the urban ecosystems that account for alternate decision pathways on different land uses.

Keywords:

Biology and life sciences – Organisms – Eukaryota – Plants – Shrubs – Trees – Conifers – Ecology – Community ecology – Community structure – Ecosystems – Urban ecosystems – Ecological metrics – Species diversity – Urban ecology – Earth sciences – Geography – Human geography – Land use – Social sciences – Human geography – Land use – Ecology and environmental sciences – Ecology – Community ecology – Community structure – Ecosystems – Urban ecosystems – Ecological metrics – Species diversity – Urban ecology

Introduction

Woody vegetation community composition, structure, and distribution are largely controlled by human decisions and actions in urban ecosystems [17]. Development, landscaping, and ongoing maintenance are important milestones for management decisions that determine vegetation community characteristics. Changes to the vegetation community alter ecosystem service provision and habitat quality and quantity [1,8,9].

Development has replaced fire as the primary disturbance driver and precursor to new forest stands in the coastal Puget Sound region of Washington [4,7,10,11]. The mechanisms of disturbance when clearing and grading land for development include removing vegetation, removing topsoil, and compacting soil with heavy equipment [1216]. Decisions made by developers and landowners at the time of development determine the extent of disturbance and influence future site conditions (Fig 1). For example, choosing to preserve existing trees determines legacy vegetation and influences stand characteristics like age and size [13].

<h2>Commercial development project located in Redmond, Washington.</h2>
Fig. 1.

Commercial development project located in Redmond, Washington.


Depicted: a. clearing the site of vegetation and b. grading the site and digging the foundation. Photo credit: K. Dyson.

Vegetation succession in urban ecosystems is determined through ecological processes such as dispersal and regeneration from seed banks and through landscaping and ongoing maintenance decisions made by developers and landowners [17]. The latter has become the dominant process [1,13,15,1821]. Ornamental introduced shrubs, trees, or grasses are often chosen for landscaping, though using native species is becoming more common [1,14,19,2224]. Once planted, these require significant ongoing maintenance inputs to arrest succession and maintain the desired aesthetic [1,17,25,26]. Along with trees retained through tree preservation policies, landscape plantings represent a significant portion of the vegetation on site and of the habitat quality and quantity available to other organisms [1,2].

Drivers determining vegetation management decisions, actions, and outcomes are multi-scalar, and include policy, community social pressures, aggregated neighborhood socio-economic status, and the motivations and preferences of individual landowners [27]. Relevant public policies include clearing and grading permitting processes, impervious surface maximums and minimums via parking space requirements, tree protection policies, canopy cover goals, and vegetation planting policies [2830]. These policies are frequently enacted to protect ecosystem services, including carbon sequestration and aesthetic preferences [13,20,3134].

Community drivers include the social norms and customs that influence individual behavior [27]. On residential properties, homeowners alter preferences for their own yards in response to the choices of nearby neighbor’s yards [35], though their assumptions about neighbor preference are not always accurate [36]. On commercial properties, owners may alter preferences to appeal to prospective and existing tenants [37,38].

Neighborhood socio-economic status is often identified an important predictor of vegetation communities in studies of residential property. These variables are aggregated to the neighborhood scale, though they reflect group membership of the individual. Group membership often serves as a proxy for commonly held attitudes and ability to manipulate their environment [39], however they are also inextricably linked with systematic forces of inequality influencing the spatial distribution of wealth in a city [40]. Socio-economic variables correlated with canopy cover and other vegetation metrics include: current and historic household income [2,39,4149], education level [21,45,48], ethnic composition [39,41,48,50], home value [51], home ownership [39], and housing age [2,44,46,47,52]. However, researchers that disaggregate socio-economic characteristics find that individual attitudes may be more important than these aggregated measures that serve as a proxy [21,53].

In municipal parks, education level and park age were only occasionally important [21,54]. These aggregated measures are thought to influence vegetation through neighborhood investment, advocacy, and legacy effects [40,44,55]. Individual scale drivers on other land uses are poorly studied.

Developers for all land uses are often motivated by cost and investment decisions [56]. Bulk construction paired with removing existing vegetation is purportedly cheaper, though preserving vegetation may be less expensive in the long run [14].

These management decisions which create vegetation communities and patterns in cities also impact ecosystem function, food webs, and biodiversity [1,2,13,14,57,58]. Different tree and shrub species have different capacity for carbon sequestration [59,60]. Introduced ornamentals generally do not same insect species, or the same biomass or diversity of fauna as native habitat [14,23,6163]. These changes to habitat quality and quantity also impact higher trophic levels [1,23,6468]. For the urban matrix to support conservation, decision makers across land uses need to take actions that support locally important vegetation habitat [69,70].

While the drivers and outcomes of decision making are increasingly well studied on residential private property, other land uses have not been given the same attention [71,72]. For example, commercial and industrial land uses are generally included only as independent variables in remote sensing studies of factors influencing percent canopy cover [51,73]. Additionally, research where the unit of analysis is defined by the area of influence of specific decision makers is also needed. Aggregated measures, such as vegetation transects through neighborhoods or canopy cover of a census block, cannot examine specific decision outcomes as they conflate different actors and their motivations and actions, and previous research shows that motivations differ between actors [21,59].

To fill this gap, I examined woody vegetation community composition on office developments in Bellevue and Redmond, Washington, USA. Specifically, I examined 1) tree and shrub communities present on office developments and 2) whether aggregated and parcel specific socio-economic variables or development and landscaping outcomes better explained observed variation in vegetation communities.

I hypothesized that vegetation communities on office developments would be heterogeneous. I also hypothesized that aggregated socio-economic variables found significant in explaining vegetation patterns on residential property would not be significant on office developments [2,34,42], but that parcel scale variables would. Finally, I hypothesized that the outcome of development and landscaping actions would better explain variation in tree and shrub community structure. I found that woody vegetation communities on office developments are heterogenous with distinct community types, and that in contrast with residential property, development and landscaping actions explain this variability better than socio-economic variables.

Materials and methods

This study was approved by the University of Washington Human Subjects Division under Determination of Exemption #48246. Field surveys were approved in writing by private property owners or managers of office developments. I did not perform any animal research or collect plant, animal, or other materials from a natural setting.

Study area and site selection

Redmond (2017 population 64,000) and Bellevue (population 144,000) are located east of Seattle in King County, Washington [74]. Both cities share a similar ecological history, a similar disturbance timeline for logging and agriculture, and have grown considerably since the opening of the Evergreen Point Floating Bridge (SR 520) in 1963. They are at similar elevations (< 160 m) and experience the same climate and similar weather.

The sampling frame was limited to Redmond and Bellevue north of I-90, excluded developments in Bellevue’s central business district, and contained parcels defined as office use by the King County Assessor’s Office (Fig 2). I grouped adjacent parcels built within three years of one another and with the same owner to create a unit of analysis based on human action not cadastral boundaries. This initial population size was 492 developments.

<h2>Map of office development study sites in Redmond and Bellevue, Washington.</h2>
Fig. 2.

Map of office development study sites in Redmond and Bellevue, Washington.


The population of office developments with High (HH), Medium Canopy (MC), Medium Diverse (MD), Medium (MM), and Low (LL) vegetation types are represented with colored circles; excluded sites (no vegetation/LP, wetlands/WW, and under construction/XX) are represented with gray triangles. Sampled sites are shown with colored diamonds.

I used disproportionate stratified random sampling to ensure that my sample included sites across the entire vegetation gradient. I classified the vegetation at each potential study site into categories using visual estimation during site visits in winter 2014 (Fig 3, Table 1). Sites with no vegetation, with wetlands, or those that were currently under construction or undergoing landscape replanting were excluded from the analysis (total 87 sites). The remaining pool of 405 potential sites had no notable hydrological features on site.

<h2>Examples of each vegetation type.</h2>
Fig. 3.

Examples of each vegetation type.


From top left to bottom right: High (HH); Medium Canopy (MC); Medium Diverse (MD); Medium (MM); Low (LL); no vegetation (LP; excluded); wetlands (WW; excluded).

Tab. 1.

Vegetation type assignment criteria and strata size.

<h2>Vegetation type assignment criteria and strata size.</h2>

I conducted stratified random sampling on sites with High, Medium Canopy, Medium Diverse, Medium, and Low vegetation types. Site selection was restricted based on site area and surrounding impervious surfaces for concurrent studies [75]. Limiting sampling of these extremes reduced my ability to detect community differences along these gradients, though socio-economic variables are not covariate.

I requested property access through three mailings sent to the property owner or manager on file in the King County Assessor’s database [76]. I targeted vegetation categories underrepresented in my sample in the second and third mailings. Of 46 mailed requests, 20 (43.5%) received no response or were not deliverable. Of the 26 (56.5%) responses received, 6 (23.1%) of were rejected and 20 (76.9%) were accepted in writing by an individual with authority to do so [76].

Commercial use of sample sites included light industrial, white collar office space, and medical/dental offices. Some sites were fully leased to tenants, while others were either partly or fully owner-occupied. Company size ranged from less than 10 to many thousand employees.

Independent variables

I examined three categories of independent variables: 1. aggregated and parcel scale socio-economic variables, 2. development and landscaping outcome variables, and 3. ground cover material and maintenance regime variables (Table 2). Aggregated socio-economic variables, sometimes called neighborhood socio-economic variables, reflect group membership of individual landowners aggregated to a common level (in the United States, often block groups). Parcel scale socio-economic variables reflect individual pieces of land, including building age and value. Socio-economic variables were derived from existing databases [7781]. Variables were chosen based on previous research and analyzed in QGIS 3.2 [42,54,8285].

Tab. 2.

Definition of independent variables used in PERMANOVA and correlation analysis [7781].

<h2>Definition of independent variables used in PERMANOVA and correlation analysis [<em class="ref">77</em>–<em class="ref">81</em>].</h2>

I measured the height of dominant native conifers using a Nikon Forestry Pro Laser Rangefinder. This is a proxy measure for age as I did not collect tree cores due to liability concerns [76]. I used historical records and site construction plans to determine whether each site retained a stand of three adjacent tree predating site development. I used Pseudotsuga menziesii (Mirb.) Franco, Thuja plicata Donn ex D. Don, and Tsuga heterophylla (Raf.) Sarg. counts to calculate native conifer density.

I digitized broad ground cover material classes in QGIS to calculate area [85]. Pervious cover types recorded include dense vegetation, dirt/litter, lawn (turf grass including moss and forb species), gravel, dense ivy, mulch, and water. I used semi-structured interviews of property owners, managers, and landscaping services along with site visits to obtain maintenance regime variables [86,87]. Irrigation, mulching, herbicide, and fertilizer application had only three “no” responses and thus could not be used to draw any well supported conclusions.

Vegetation data collection

I censused vegetation communities during the summer of 2015, excluding saplings with DBH < 3”. Each tree or shrub was identified to species or genus in consultation with experts at the Center for Urban Horticulture at University of Washington [8890]. Some tree and shrub species were grouped at the genus level due to the abundance of very similar cultivars in the landscaping trade, including Malus Mill. [46]. Following previous studies, I grouped conifers under 2 m into a broad class of dwarf conifer species [91]. 10 individual trees (0.506% of total trees) and 14 shrubs (0.174% of total shrubs) could not be identified; these were given a unique identifier code for multivariate community analysis.

I assigned tree and shrub genera to one of three provenance categories—native, non-native, or ambiguous [92,93]. The ambiguous category was used for genera including both native and non-native cultivated species that are difficult to distinguish, and/or frequently interbred and sold as crosses. For example, some Mahonia Nutt. species are native (M. aquifolium Pursh Nutt. and M. nervosa Pursh Nutt.), while others originate in Asia (Mahonia japonica Thumb. DC.) and many hybrids are bred and sold by nurseries (e.g. Mahonia x media “Charity” Brickell).

Identifying and describing vegetation clusters on office developments

I standardized tree and shrub abundance data and ground cover area by total site area in acres. This transformation preserves parcel boundaries as the unit of analysis and reflects developer and landowner actions during and following development that determine the amount of impervious surface and pervious area, the number of trees preserved, and the number of trees and shrubs planted. All analyses were performed in R [94].

To delineate vegetation community clusters on office developments, I used a flexible agglomerative nesting function (agnes {vegan}) with a beta of -0.5 to produce an ecologically interpretable dendrogram with minimal chaining [9599]. Using the resulting groups, I performed indicator species analysis, which assesses the predictive values of species as indicators of the conditions at site groups, using multipatt {indicspecies} [100102] and a custom wrapper for repetition of the permutation-based function [103]. To examine citywide patterns, I extrapolated cluster membership to the entire population of office developments in the study area using proportions.

I used simple univariate permutational analysis of variance (PERMANOVA) models to test if continuous variables differed between vegetation community cluster groups and Pearson’s Chi-squared test to test if categorical variables differed [99,104]. PERMANOVA is a permutation-based implementation of analysis of variance (ANOVA) that avoids assumptions about underlying distributions of community structure and can be used with non-Euclidian distance matrices [104]. Bartlett tests of homogeneity found no difference between group variances (bartlett.test {stats}).

Explaining variation in tree and shrub community structure

I analyzed tree and shrub communities separately to determine if the two communities responded differently, and as development and landscaping outcome variables were derived from measurements of the tree community.

I used non-metric multidimensional scaling (NMDS) to evaluate relationships between development and landscaping variables and the tree community [95,99]. NMDS is a rank-based ordination technique that is robust to data without identifiable distributions, can be used with any distance or dissimilarity measure [95]. To determine the relationship between development and landscaping outcome variables and the tree community, I used convex hull plots and fitted environmental vectors [99].

I used PERMANOVA to test relationships between other variables and the tree community and all variables and the shrub community. I used a multi-step approach to avoid transforming independent variables or using ordination to collapse related variables, as these actions make results less interpretable for urban planners and other professionals. I first tested each independent variable with community matrices in a simple multivariate PERMANOVA model. To ensure differences in categorical variables were due to location and not dispersion, I used ANOVA to test for significant differences in dispersion [99]. I then constructed models using all variables with significant pseudo-F values in all possible single and multiple multivariate model combinations. Significance was assessed at the α ≤ 0.05 level following Holm-Bonferroni correction for multiple comparisons. I used a custom Akaike information criterion with correction (AICc) function based on residual sums of squares to compare models and identify those with the best support [103].

Results and discussion

Observed woody vegetation communities

I recorded a total of 1,978 trees and 8,039 shrubs from 52 and 84 taxonomic groups respectively (S1 and S2 Tables). Only Rhododendron L. were found on all 20 sites surveyed. Four tree species and nine shrub species were found on more than half of all office developments, with 23 tree species and 30 shrub taxa found only on only one development.

Native tree species accounted for 68.1% of total individuals observed, and three of the top five most abundant species. On average, native species accounted for 63.4% of the trees found on each office development, though sites varied widely with 0%–99% native trees. Pseudotsuga menziesii was by far the most abundant tree species, with 37.7% of observed individuals. Thuja plicata (12.4%), Acer macrophyllum Pursh (11%), Acer rubrum L. (6.7%), and Acer platanoides L. (5.1%) complete the top five. Prunus L. and Alnus rubra Bong. were both widespread taxa (found on 12 and 9 sites, respectively) but were never abundant on any one site.

In contrast, native shrub species accounted for only 30.4% individual shrubs observed. On average, native shrubs accounted for 26% of the shrubs observed at each office development, and never more than 63.2% of individual shrubs. The two most abundant shrub species were the native Gaultheria shallon Pursh (15.8%), which frequently occurs in low, dense mats, and Berberis Mahonia gp. Nutt. (12.5%) which is comprised of native, introduced, and hybrid species. The rest of the top five most abundant shrub species were all non-native, including Prunus laurocerasus L. (8.5%), Rhododendron (7.6%), and Cornus sericea L. (5.2%).

Measures of tree and shrub abundance, density, and diversity varied substantially between sites (Table 3). In general, total species richness and native species richness were positively correlated (Pearson’s Correlation for Tree: 0.594; Shrub: 0.545), though four sites with above average species richness had three or fewer native species planted. Remnant large native conifer abundance, primarily Pseudotsuga menziesii, greatly contributed to sites with greater tree abundance (Pearson’s: 0.83); consequently, Shannon diversity was generally lower on sites with more native trees (Pearson’s: -0.407).

Tab. 3.

Metrics for tree and shrub communities on sampled office developments.

<h2>Metrics for tree and shrub communities on sampled office developments.</h2>

Overall, these measures are within the ranges reported by other urban ecology studies, though differences in methodology and particularly the use of small plots [47] and remote sensing [48] in other studies and stratified sampling along a vegetation gradient in this study make comparison more difficult. The most abundant tree species on office developments are similar to those on residential properties in western Washington [51,59]. Measures of diversity were generally lower than residential property [47,54]. Species richness was comparable to other commercial land uses and city parks [47,54] though lower than residential land uses [46,47,52,54]. Measures of beta diversity, suggesting low similarity between locations, were also comparable [46].

Divergent vegetation groups found on office developments

I identified two groups of tree and shrub vegetation (flexible beta = -0.5; agglomerative coefficients of 0.87 and 0.76 respectively; Table 4). Using indicator species analysis, I identified the Native Tree group (11 sites) as characterized by Thuja plicata, Acer macrophyllum Pursh, Arbutus menziesii Pursh, and Alnus rubra Bong, while the Ornamental Tree group (9 sites) is characterized by Acer rubrum L. The Native Shrub group (11 sites) is characterized by Gaultheria shallon Pursh, Mahonia gp. Nutt., Symphoricarpos Duham., and Ribes sanguineum Pursh, and the Ornamental Shrub group (9 sites) by Thuja occidentalis L.

Tab. 4.

Rank abundance of tree and shrub taxa for each community group identified by flexible-beta cluster analysis.

<h2>Rank abundance of tree and shrub taxa for each community group identified by flexible-beta cluster analysis.</h2>

The two groups are distinct in the average density of trees and shrubs per site (Native Tree mean = 58, Ornamental Tree mean = 25.7 with Pr(>F) = 0.003; Native Shrub mean = 226.6, Ornamental Shrub mean = 92.9 with Pr(>F) = 0.001). The mean median height of dominant native conifers was also significantly different between clusters for trees and shrubs (Native Tree mean = 33.2 m, and Ornamental mean = 16.8 m, with Pr(>F) = 0.001; Native Shrub mean = 32.6 m, and Ornamental mean = 20.2 m, with Pr(>F) = 0.03). However, there was no difference in area between Native and Ornamental clusters for either trees or shrubs (tree Pr(>F) = 0.424; shrub Pr(>F) = 0.599). Dead wood abundance was significantly greater on Native Tree sites than Ornamental Tree sites (Native Tree mean = 13.4, and Ornamental mean = 2.4, with Pr(>F) = 0.019), but not between shrub groups.

Only impervious surface cover between Native and Ornamental Tree sites differed significantly (Native Tree mean = 60, and Ornamental mean = 70, with Pr(>F) = 0.007). No other ground covers differed.

There was also substantial co-occurrence between Native and Ornamental groups. Of the 20 office developments surveyed, nine sites belong to both Native Tree and Shrub community groups, and seven sites belong to both Ornamental Tree and Shrub community groups. This suggests that the sequential decisions made concerning tree preservation, tree plantings, and shrub plantings are related. The observed differences in species composition, between group differences, and high turnover (beta diversity) support the conclusion that woody vegetation communities on office developments are heterogenous.

Native Tree and Shrub communities are more rare than Ornamental Tree and Shrub communities. Extrapolation suggests there are approximately 70 Native Tree and 335 Ornamental Tree developments (17.3%), and 152 Native Shrub and 253 Ornamental Shrub developments (37.5%). The accuracy of these estimates is influenced by the Medium vegetation type, as it is large and proportionally under sampled, and the relatively small sample size.

Socio-economic variables poorly explain variation in tree or shrub community composition

Neither aggregated neighborhood scale measures of residential socio-economic status nor parcel scale measures of economic value and the built environment explained variation in tree and shrub community composition on office developments following Holm-Bonferroni correction (S3 Table). For the tree community, median household income was significant only before correction. This largely supports my hypothesis that aggregated socio-economic variables describing neighborhoods are not important for adjacent commercial properties as well [41,42], though additional research is needed.

Theoretically, for residential socio-economic variables to drive vegetation on office developments, the adjacent residential context must influence developer and commercial landowner vegetation choices. Generally, zoning code in Bellevue and Redmond seeks to screen land uses from one another. Owners of office developments are likely signaling to prospective and existing tenants [37,38], in contrast with owners of residential properties, who use vegetation choices to signal to their neighbors of similar socio-economic status [27,35,36]. Other explanations include proximity to desirable amenities–that is, both residential and commercial properties near amenities are appealing to more wealthy neighbors/tenants, the influence of city design review boards, or neighborhood overlay districts.

Studies examining why decision makers on commercial property make planting decisions are fewer in number than residential homeowners, though existing studies provide important early insight. For example, for landscape architects in Toronto factors like site aspect, appearance, and available space rated more highly in species selection than whether species are native or nearby canopy composition [34].

Parcel scale measures of economic value and the built environment were not significant, which fails to support my hypothesis. Previous research found that property value explained variation in the woody vegetation community [51]. Similarly, site age was suggested as a determinant of woody vegetation community composition by studies on residential properties [2,44], landscaping professionals I interviewed, and my examination of contemporaneous landscaping plans filed with the cities of Bellevue and Redmond. Landscaping professionals mentioned trends in plant popularity, including Pieris japonica (Thunb.) D. Don ex G. Don in the late 1980s and increasing use of native plants like Ribes sanguineum since 2000. Alternative explanations for this finding include that building age is a poor measure for landscaping age due to replanting; an interaction between age and landscaping budget; or that a subset of office developments are planted with in vogue landscape plants, such as the common but under sampled Medium vegetation type.

Differences in study design may also be responsible for these divergent results. Other studies use index response variables and univariate regression [42, 54], measures dependent on effort [54, 107], and plot or transect designs which confound different actors and outcomes [47, 72].

Development and landscaping outcomes are related to tree and shrub community composition

Multiple variables describing the outcome of development and landscaping actions explain variation in tree and shrub community composition. For the tree community, NMDS with convex hulls and fitted environmental vectors found strong relationships with median dominant native conifer height, native conifer density, the presence of stands predating development, and dead wood abundance (particularly stump abundance, Fig 4). These variables were also included in the best supported PERMANOVA models for the shrub community (Table 5).

<h2>Two dimensional NMDS representation of tree community composition.</h2>
Fig. 4.

Two dimensional NMDS representation of tree community composition.


Median dominant native conifer height, native conifer density, and the presence of stands predating development are associated with the first NMDS axis. Dead wood is associated with both axes. Black dots represent sites with stands predating development, gray dots sites without. Ordination has not been rotated prior to plotting.

Tab. 5.

PERMANOVA model summary comparing multivariate models of shrub community composition.

<h2>PERMANOVA model summary comparing multivariate models of shrub community composition.</h2>

Together, these results support my hypothesis that development and landscaping actions impact vegetation communities on office developments. They also agree with some residential researchers who found that homeowner attitudes and actions were more important than socio-economic descriptors [53]. Together with flexible beta clustering results, this suggests that for each development a suite of decisions is made that results in either retaining more trees and planting native shrubs or retaining fewer trees and planting ornamental trees and shrubs.

However, development and landscaping outcomes are the end point of economic decision making processes poorly studied in urban ecology. Though the coarse socio-economic variables examined here were not significant, developer and landowner motivations and decision making were not considered explicitly, only their outcomes. To reach these end points, developers may consider ease of construction based on site conditions, relative cost of different construction approaches, preferences of the landowner and customer specifications, previous company experience or company aesthetic, and development regulations [13,34,108110]. The intended audience of prospective and existing tenants may influence both development and landscaping decisions [37,38]. These considerations may influence financing available to developers, financial risk, and the appeal of and thus demand for the completed project [37]. Further, when considering multiple competing options—such as different landscaping choices—developers and landowners may satisfice [111]. That is, they search through alternatives until one meets an acceptability threshold and choose that option.

Implications for urban habitat quality and quantity

The woody vegetation communities I observed on office developments suggest that integrating habitat conservation during and following development is possible using currently existing developments as models. As local biological communities are largely determined by vegetation, sites with more trees preserved and a greater abundance of native conifers likely provide higher habitat quality and quantity to other organisms [1,2,14,57,58] as native vegetation is more likely to support native insects and native birds than ornamental plantings [23,75,112116]. One estimate suggests native vegetation volume must be above 70% in order to maintain populations of native insectivorous bird species [116]; sites with high numbers of trees preserved may already hit this target.

We can point to actions and policies more likely to support high quality habitat and benefit other trophic levels, including tree preservation policies, promoting native tree and shrub planting, and removing policy barriers to native vegetation [75,117,118]. However, the motivations driving exemplary adoption of these actions are currently opaque. Anecdotes shared during fieldwork suggest owner-occupied office space, cost, and personal values and connections to nature may be important factors in determining development and landscaping actions, as with homeowners [21,36,119123].

Implications for future urban ecology research

Observed within land use heterogeneity and between land use differences in socio-economic variable importance both have implications for urban ecology research. Within land use heterogeneity results in vegetation distributions that are non-normal, with likely kurtosis and heteroscedasticity (e.g. Fig 5). Therefore, choice of sampling design and statistical method can result in inaccurate conclusions, particularly in conjunction with small sample size [124]. Sampling across important gradients, as with vegetation composition in this study, is particularly important. Additionally, potential solutions already extant on the landscape may be overlooked. This provides support for stratified sampling designs, larger sample sizes, and choosing analysis methods robust to broken assumptions of normality of the sampled population [125].

<h2>Hypothesized distribution of the number of trees on office developments based on observed mean and standard deviations for each vegetation class used in sampling (HH, MC, MD, MM, LL).</h2>
Fig. 5.

Hypothesized distribution of the number of trees on office developments based on observed mean and standard deviations for each vegetation class used in sampling (HH, MC, MD, MM, LL).


Note heavy right tail from HH,MC, and MD sites (kurtosis); each vegetation class also has a different variance (heteroscedasticity).

Researchers should choose their sampling strategy carefully based on research questions and the underlying distribution of key variables in urban contexts with long environmental gradients [124,126,127]. If the phenomenon of interest is related to the vegetation community, researchers should attempt to better understand and sample the vegetation gradient (e.g. via stratified sampling) instead of sampling only along a measure of the built gradient [128]. Here, collecting vegetation information for the population of office developments prior to sampling improved sampling design.

Between land use differences in socio-economic variable importance suggests that creating vegetation models of land use within a city is likely inaccurate if all land uses are assumed to respond equivalently to a given variable. Researchers cannot assume that vegetation gradients and socio-economic gradients are parallel; these gradients may also interact resulting in heteroscedasticity. Decision pathways to support carbon sequestration and habitat models need to be constructed based on research for each land use separately.

Conclusion

Humans control woody vegetation communities in urban ecosystems, influencing ecosystem service provision and habitat quality and quantity. Commercial land uses, including office developments, have largely been overlooked in studies of urban woody vegetation composition and studies examining how these communities are assembled. I filled this gap by examining tree and shrub communities on office developments in Redmond and Bellevue, Washington, USA.

I found that the vegetation communities on these developments are heterogenous, with distinct groups of sites characterized by Native and Ornamental Tree and Shrub vegetation communities. I also found that socioeconomic measures aggregated to the neighborhood scale and at the parcel scale were less important in explaining variation in community composition than variables describing the outcomes of development and landscaping choices.

This research contributes to our understanding of vegetation communities outside of municipal parks and residential land uses. It is also one of few studies that uses site surveys where the unit of measurement is based on how management decisions are made, instead of methods derived from wildlands vegetation research [47,72] or remote sensing [48].

The observed heterogeneity in vegetation communities suggests that different ecosystem functions and habitat quantity and quality are provided on office developments. Greater provision of these functions is possible using currently existing developments as models. Further, within land use heterogeneity suggests that urban ecology research must more carefully consider sampling design, particularly sampling along key environmental gradients. The observed differences in variable importance between office developments and residential land uses suggests that future research and models of the urban ecosystem must account for land uses’ different decision pathways.

Going forward, researchers should examine other commercial land uses, commercial land use in additional ecotypes, and particularly the decision pathways followed by actors on commercial and other land uses. This research agrees with other studies suggesting that specific actions are more important than aggregated socio-economic variables [53]. Additional research is needed to link decision makers’ personal values and aesthetic preferences, economic motivations, and social norms with tree and shrub community composition on commercial land following work on residential property [27,53]. Needed studies include interviews to better understand tree preservation and planting motivations [34,109]; aesthetic preference studies as on residential developments [129,130]; and tracing decision making pathways based on previous land use [131]. A better understanding of these processes may improve habitat quality and quantity on commercial property [132]. Finally, research is also needed to determine if vegetation inequity observed on residential properties [39] is perpetuated on commercial properties. The No Vegetation type excluded from analysis here was often adjacent to retail use, where worker compensation is generally less than in medical/dental, software, and other white collar jobs in office developments.

Supporting information

S1 Table [docx]
All trees observed in site surveys.

S2 Table [docx]
All shrubs observed in site surveys.

S3 Table [docx]
All simple multivariate PERMANOVA results for tree and shrub communities.


Zdroje

1. Faeth SH, Bang C, Saari S. Urban biodiversity: Patterns and mechanisms. Annals of the New York Academy of Sciences. 2011;1223(1):69–81.

2. Avolio ML, Pataki DE, Trammell TL, Endter-Wada J. Biodiverse cities: The nursery industry, homeowners, and neighborhood differences drive urban tree composition. Ecological Monographs. 2018;88(2):259–76.

3. Pickett ST, Cadenasso ML, Grove JM, Nilon CH, Pouyat RV, Zipperer WC, et al. Urban ecological systems: Linking terrestrial ecological, physical, and socioeconomic components of metropolitan areas. In: Urban Ecology. Springer; 2008. pp. 99–122.

4. Gibb H, Hochuli DF. Habitat fragmentation in an urban environment: Large and small fragments support different arthropod assemblages. Biological Conservation. 2002;106(1):91–100.

5. Mullaney J, Lucke T, Trueman SJ. A review of benefits and challenges in growing street trees in paved urban environments. Landscape and Urban Planning. 2015;134:157–66.

6. Peters DP, Lugo AE, Chapin FS, Pickett ST, Duniway M, Rocha AV, et al. Cross-system comparisons elucidate disturbance complexities and generalities. Ecosphere. 2011;2(7):1–26.

7. Sharpe DM, Stearns F, Leitner LA, Dorney JR. Fate of natural vegetation during urban development of rural landscapes in southeastern Wisconsin. Urban Ecology. 1986;9(3–4):267–87.

8. Byrne LB. Habitat structure: A fundamental concept and framework for urban soil ecology Urban Ecosystems. 2007 Sep;10(3):255–74.

9. Lehmann I, Mathey J, Rößler S, Bräuer A, Goldberg V. Urban vegetation structure types as a methodological approach for identifying ecosystem services–Application to the analysis of micro-climatic effects. Ecological Indicators. 2014;42:58–72.

10. Walcott CD. Nineteenth annual report of the united states geological survey to the secretary of the interior 1897–1898: Part v–forest reserves [Internet]. 1899. https://pubs.er.usgs.gov/publication/ar19_5

11. Halpern CB, Spies TA. Plant species diversity in natural and managed forests of the pacific northwest. Ecological Applications. 1995;5(4):913–34.

12. Andres CK, Smith RC. Principles and practices of commercial construction. Pearson/Prentice Hall; 2004.

13. Dorney JR, Guntenspergen GR, Keough JR, Stearns F. Composition and structure of an urban woody plant community. Urban Ecology. 1984;8(1–2):69–90.

14. McKinney ML. Urbanization, biodiversity, and conservation the impacts of urbanization on native species are poorly studied, but educating a highly urbanized human population about these impacts can greatly improve species conservation in all ecosystems. BioScience. 2002;52(10):883–90.

15. Grimm NB, Pickett ST, Hale RL, Cadenasso ML. Does the ecological concept of disturbance have utility in urban social–ecological–technological systems? Ecosystem Health and Sustainability. 2017;3(1):e01255.

16. Turner MG. Landscape ecology: What is the state of the science? Annu Rev Ecol Evol Syst. 2005;36:319–44.

17. Zipperer WC. The process of natural succession in urban areas. The Routledge Handbook of Urban Ecology. 2010;187.

18. Widrlechner MP. Trends influencing the introduction of new landscape plants. Advances in new crops Timber Press, Portland, OR. 1990;460–7.

19. van Heezik YM, Freeman C, Porter S, Dickinson KJ, others. Native and exotic woody vegetation communities in domestic gardens in relation to social and environmental factors. Ecology and Society. 2014;19(4):17.

20. Goodness J. Urban landscaping choices and people’s selection of plant traits in cape town, south africa. Environmental Science & Policy. 2018;85:182–92.

21. Kendal D, Williams KJ, Williams NS. Plant traits link people’s plant preferences to the composition of their gardens. Landscape and Urban Planning. 2012;105(1–2):34–42.

22. Germaine SS, Rosenstock SS, Schweinsburg RE, Richardson WS. Relationships among breeding birds, habitat, and residential development in greater Tucson, Arizona. Ecological applications. 1998;8(3):680–91.

23. Burghardt KT, Tallamy DW, Gregory Shriver W. Impact of native plants on bird and butterfly biodiversity in suburban landscapes. Conservation Biology. 2009;23(1):219–24. doi: 10.1111/j.1523-1739.2008.01076.x 18950471

24. Blair RB. Land use and avian species diversity along an urban gradient. Ecological Applications [Internet]. 1996;6(2):506–19. Available from: http://dx.doi.org/10.2307/2269387

25. LeBauer DS, Treseder KK. Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology. 2008;89(2):371–9. doi: 10.1890/06-2057.1 18409427

26. Lepczyk CA, Mertig AG, Liu J. Assessing landowner activities related to birds across rural-to-urban landscapes. Environmental Management. 2004;33(1):110–25. doi: 10.1007/s00267-003-0036-z 14749899

27. Cook EM, Hall SJ, Larson KL. Residential landscapes as social-ecological systems: A synthesis of multi-scalar interactions between people and their home environment. Urban Ecosystems. 2012;15(1):19–52.

28. Young RF. Planting the living city: Best practices in planning green infrastructure—Results from major US cities. Journal of the American Planning Association. 2011;77(4):368–81.

29. Environmental Protection Agency. Assessing street and parking design standards to reduce excess impervious cover in new hampshire and massachusetts. 2011.

30. DeLaria M. Low impact development as a stormwater management technique. The Rocky Mountain Land Use Institute; 2008.

31. Wolf KL. Business district streetscapes, trees, and consumer response. Journal of Forestry. 2005;103(8):396–400.

32. Collins SL, Carpenter SR, Swinton SM, Orenstein DE, Childers DL, Gragson TL, et al. An integrated conceptual framework for long-term social–ecological research. Frontiers in Ecology and the Environment. 2011;9(6):351–7.

33. Elmendorf W. The importance of trees and nature in community: A review of the relative literature. Arboriculture and Urban Forestry. 2008;34(3):152.

34. Conway TM. Tending their urban forest: Residents’ motivations for tree planting and removal. Urban forestry & urban greening. 2016;17:23–32.

35. Nassauer JI, Wang Z, Dayrell E. What will the neighbors think? Cultural norms and ecological design. Landscape and Urban Planning. 2009;92(3–4):282–92.

36. Peterson MN, Thurmond B, Mchale M, Rodriguez S, Bondell HD, Cook M. Predicting native plant landscaping preferences in urban areas. Sustainable Cities and Society. 2012;5:70–6.

37. Laverne RJ, Winson-Geideman K, others. The influence of trees and landscaping on rental rates at office buildings. Journal of Arboriculture. 2003;29(5):281–90.

38. Levy D, Peterson G. The effect of sustainability on commercial occupiers’ building choice. Journal of Property Investment & Finance. 2013;31(3):267–84.

39. Heynen N, Perkins HA, Roy P. The political ecology of uneven urban green space: The impact of political economy on race and ethnicity in producing environmental inequality in Milwaukee. Urban Affairs Review. 2006;42(1):3–25.

40. Schell CJ, Dyson K, Fuentes TL, Lambert MR. The ecological consequences of social inequality. 2019.

41. Leong M, Dunn RR, Trautwein MD. Biodiversity and socioeconomics in the city: A review of the luxury effect. Biology Letters. 2018;14(5):20180082. doi: 10.1098/rsbl.2018.0082 29743266

42. Hope D, Gries C, Zhu W, Fagan WF, Redman CL, Grimm NB, et al. Socioeconomics drive urban plant diversity. Proceedings of the national academy of sciences. 2003;100(15):8788–92.

43. Larsen L, Harlan SL. Desert dreamscapes: Residential landscape preference and behavior. Landscape and urban planning. 2006;78(1–2):85–100.

44. Boone CG, Cadenasso ML, Grove JM, Schwarz K, Buckley GL. Landscape, vegetation characteristics, and group identity in an urban and suburban watershed: Why the 60s matter. Urban Ecosystems. 2010;13(3):255–71.

45. Krafft J, Fryd O. Spatiotemporal patterns of tree canopy cover and socioeconomics in Melbourne. Urban Forestry & Urban Greening. 2016;15:45–52.

46. Sierra-Guerrero MC, Amarillo-Suárez AR. Socioecological features of plant diversity in domestic gardens in the city of Bogotá, Colombia. Urban Forestry & Urban Greening. 2017;28:54–62.

47. Clarke LW, Jenerette GD, Davila A. The luxury of vegetation and the legacy of tree biodiversity in Los Angeles, ca. Landscape and Urban Planning. 2013;116:48–59.

48. Luck GW, Smallbone LT, O’Brien R. Socio-economics and vegetation change in urban ecosystems: Patterns in space and time. Ecosystems. 2009;12(4):604.

49. Avolio ML, Pataki DE, Pincetl S, Gillespie TW, Jenerette GD, McCarthy HR. Understanding preferences for tree attributes: The relative effects of socio-economic and local environmental factors. Urban Ecosystems. 2015;18(1):73–86.

50. Grove JM, Cadenasso ML, Burch WR Jr, Pickett ST, Schwarz K, O’Neil-Dunne J, et al. Data and methods comparing social structure and vegetation structure of urban neighborhoods in Baltimore, Maryland. Society and Natural Resources. 2006;19(2):117–36.

51. Mills JR, Cunningham P, Donovan GH. Urban forests and social inequality in the pacific northwest. Urban Forestry & Urban Greening. 2016;16:188–96.

52. Jim CY. Trees and landscape of a suburban residential neighbourhood in Hong Kong. Landscape and Urban Planning. 1993;23(2):119–43.

53. Shakeel T, Conway TM. Individual households and their trees: Fine-scale characteristics shaping urban forests. Urban Forestry & Urban Greening. 2014;13(1):136–44.

54. Martin CA, Warren PS, Kinzig AP. Neighborhood socioeconomic status is a useful predictor of perennial landscape vegetation in residential neighborhoods and embedded small parks of Phoenix, AZ. Landscape and Urban Planning. 2004;69(4):355–68.

55. Rigolon A, Browning M, Jennings V. Inequities in the quality of urban park systems: An environmental justice investigation of cities in the united states. Landscape and Urban Planning. 2018;178:156–69.

56. Almagor J. Possible urban futures: The impact of planners and developers on urban dynamics [PhD thesis]. Tel Aviv University; 2017.

57. Faeth SH, Warren PS, Shochat E, Marussich WA. Trophic dynamics in urban communities. BioScience. 2005;55(5):399–407.

58. Wittig R. Biodiversity of urban-industrial areas and its evaluation–a critical review. Urban biodiversity and design. 2010;37–55.

59. Tenneson K. The residential urban forest: Linking structure, function and management [PhD thesis]. University of Washington; 2013.

60. Tang Y, Chen A, Zhao S. Carbon storage and sequestration of urban street trees in Beijing, China. Frontiers in Ecology and Evolution. 2016;4:53.

61. Crisp PN, Dickinson K, Gibbs G. Does native invertebrate diversity reflect native plant diversity? A case study from New Zealand and implications for conservation. Biological Conservation. 1998;83(2):209–20.

62. Rebele F. Urban ecology and special features of urban ecosystems. Global ecology and biogeography letters. 1994;173–87.

63. Mach BM, Potter DA. Quantifying bee assemblages and attractiveness of flowering woody landscape plants for urban pollinator conservation. PLOS ONE. 2018;13(12):e0208428. doi: 10.1371/journal.pone.0208428 30586408

64. Marzluff JM, Bowman R, Donnelly R. A historical perspective on urban bird research: Trends, terms, and approaches. In: Avian ecology and conservation in an urbanizing world. Springer; 2001. pp. 1–17.

65. Alberti M, Marzluff JM, Shulenberger E, Bradley G, Ryan C, Zumbrunnen C. Integrating humans into ecology: Opportunities and challenges for studying urban ecosystems. AIBS Bulletin. 2003;53(12):1169–79.

66. Alberti M. The effects of urban patterns on ecosystem function. International Regional Science Review. 2005;28(2):168–92.

67. Polasky S, Nelson E, Lonsdorf E, Fackler P, Starfield A. Conserving species in a working landscape: Land use with biological and economic objectives. Ecological applications. 2005;15(4):1387–401.

68. Rosenzweig ML. Reconciliation ecology and the future of species diversity. Oryx. 2003;37(2):194–205.

69. Goddard MA, Dougill AJ, Benton TG. Scaling up from gardens: Biodiversity conservation in urban environments. Trends in ecology & evolution. 2010;25(2):90–8.

70. Miller JR, Hobbs RJ. Conservation where people live and work. Conservation biology. 2002;16(2):330–7.

71. Snep RP, Wallis DeVries MF, Opdam P. Conservation where people work: A role for business districts and industrial areas in enhancing endangered butterfly populations? Landscape and Urban Planning. 2011;103(1):94–101.

72. Bourne KS, Conway TM. The influence of land use type and municipal context on urban tree species diversity. Urban ecosystems. 2014;17(1):329–48.

73. Fan C, Johnston M, Darling L, Scott L, Liao FH. Land use and socio-economic determinants of urban forest structure and diversity. Landscape and urban planning. 2019;181:10–21.

74. United States Census Bureau. Population and housing unit estimates [Internet]. 2017. https://www.census.gov/programs-surveys/popest.html?intcmp=serp

75. Dyson K. Parcel-scale development and landscaping actions affect vegetation, bird, and fungal communities on office developments [PhD thesis]. 2019.

76. Dyson K, Ziter C, Fuentes TL, Patterson M. Conducting urban ecology research on private property: Advice for new urban ecologists. Journal of Urban Ecology. 2019;5(1):juz001.

77. United States Census Bureau. American community survey 5yr block group [Internet]. 2016. http://census.gov/programs-surveys/acs/data.html

78. King County Department of Assessments. King county assessments data [Internet]. 2014. http://info.kingcounty.gov/assessor/DataDownload/default.aspx

79. King County GIS Center. King county gis data portal [Internet]. 2014. http://www5.kingcounty.gov/gisdataportal/Default.aspx

80. Xian G, Homer C, Dewitz J, Fry J, Hossain N, Wickham J. Change of impervious surface area between 2001 and 2006 in the conterminous united states. Photogrammetric Engineering and Remote Sensing. 2011;77(8):758–62.

81. Homer CG, Dewitz JA, Yang L, Jin S, Danielson P, Xian G, et al. Completion of the 2011 national land cover database for the conterminous united states-representing a decade of land cover change information. Photogramm Eng Remote Sens. 2015;81(5):345–54.

82. Walker JS, Grimm NB, Briggs JM, Gries C, Dugan L. Effects of urbanization on plant species diversity in central Arizona. Frontiers in Ecology and the Environment. 2009;7(9):465–70.

83. Dana E, Vivas S, Mota J. Urban vegetation of almeria city—a contribution to urban ecology in spain. Landscape and Urban Planning. 2002;59(4):203–16.

84. Grove JM, Locke DH, O’Neil-Dunne JP. An ecology of prestige in new york city: Examining the relationships among population density, socio-economic status, group identity, and residential canopy cover. Environmental management. 2014;54(3):402–19. doi: 10.1007/s00267-014-0310-2 25034751

85. QGIS Development Team. QGIS geographic information system [Internet]. Open Source Geospatial Foundation; 2016. http://qgis.osgeo.org

86. Dexter L. Elite and specialized interviewing. 1970;

87. Harvey WS. Strategies for conducting elite interviews. Qualitative Research. 2011 Aug;11(4):431–41.

88. Sibley D, others. Sibley guide to trees. Alfred A. Knopf; Distributed by Random House; 2009.

89. Dirr M. Dirr’s hardy trees and shrubs: An illustrated encyclopedia. Timber Press, Inc. 1997.

90. Dirr M. Manual of woody landscape plants: Their identification, ornamental characteristics, culture, propagation and uses. Stipes Publishing LLC; 2009.

91. Daniels G, Kirkpatrick J. Comparing the characteristics of front and back domestic gardens in Hobart, Tasmania, Australia. Landscape and Urban Planning. 2006;78(4):344–52.

92. U.S. Geological Survey. Digital representation of “Atlas of United States trees” by Elbert L. Little, Jr. [Internet]. 1999. http://gec.cr.usgs.gov/data/little/

93. USDA. The plants database [Internet]. 2016. http://plants.usda.gov/java/

94. R Core Team. R: A language and environment for statistical computing [Internet]. Vienna, Austria: R Foundation for Statistical Computing; 2019. https://www.R-project.org

95. McCune B, Grace JB, Urban DL. Analysis of ecological communities. Vol. 28. MjM software design Gleneden Beach, OR; 2002.

96. Dufrêne M, Legendre P. Species assemblages and indicator species: The need for a flexible asymmetrical approach. Ecological monographs. 1997;67(3):345–66.

97. Milligan GW. A Study of the Beta-Flexible Clustering Method. Multivariate Behavioral Research. 1989 Apr;24(2):163–76. doi: 10.1207/s15327906mbr2402_2 26755277

98. Breckenridge JN. Validating cluster analysis: Consistent replication and symmetry. Multivariate Behavioral Research. 2000;35(2):261–85. doi: 10.1207/S15327906MBR3502_5 26754085

99. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. Vegan: Community ecology package [Internet]. 2017. https://CRAN.R-project.org/package=vegan

100. De Cáceres M, Legendre P, Moretti M. Improving indicator species analysis by combining groups of sites. Oikos. 2010;119(10):1674–84.

101. De Cáceres M, Legendre P. Associations between species and groups of sites: Indices and statistical inference [Internet]. Ecology. 2009. http://sites.google.com/site/miqueldecaceres/

102. De Cáceres M. How to use the indicspecies package (ver 1.7.1). Catalonia, Centre Tecnològic Forestal de Catalunya. 2013;

103. Dyson K. Custom community ecology helper R scripts [Internet]. 2018. https://github.com/kdyson/R_Scripts

104. Anderson MJ. A new method for non-parametric multivariate analysis of variance. Austral Ecology. 2001;26(1):32–46.

105. Shannon CE, Weaver W. The mathematical theory of communication. University of Illinois Press, Urbana IL; 1949.

106. Jost L. Entropy and diversity. Oikos. 2006;113(2):363–75.

107. Karlik JF, Winer AM. Plant species composition, calculated leaf masses and estimated biogenic emissions of urban landscape types from a field survey in Phoenix, Arizona. Landscape and Urban Planning. 2001;53(1–4):123–34.

108. Grimes A, Mitchell I. Impacts of planning rules, regulations, uncertainty and delay on residential property development. 2015;

109. Häkkinen T, Belloni K. Barriers and drivers for sustainable building. Building Research & Information. 2011;39(3):239–55.

110. Nappi-Choulet I. The role and behaviour of commercial property investors and developers in French urban regeneration: The experience of the Paris region. Urban Studies. 2006;43(9):1511–35.

111. Mohamed R. Why do residential developers prefer large exurban lots? Infrastructure costs and exurban development. Environment and Planning B: Planning and Design. 2009;36(1):12–29.

112. Belaire JA, Whelan CJ, Minor ES. Having our yards and sharing them too: The collective effects of yards on native bird species in an urban landscape. Ecological Applications. 2014;24(8):2132–43. 29188686

113. Chong KY, Teo S, Kurukulasuriya B, Chung YF, Rajathurai S, Tan HTW. Not all green is as good: Different effects of the natural and cultivated components of urban vegetation on bird and butterfly diversity. Biological Conservation. 2014;171:299–309.

114. Pennington DN, Blair RB. Habitat selection of breeding riparian birds in an urban environment: Untangling the relative importance of biophysical elements and spatial scale. Diversity and Distributions. 2011;17(3):506–18.

115. Paker Y, Yom-Tov Y, Alon-Mozes T, Barnea A. The effect of plant richness and urban garden structure on bird species richness, diversity and community structure. Landscape and Urban Planning. 2014;122:186–95.

116. Narango DL, Tallamy DW, Marra PP. Nonnative plants reduce population growth of an insectivorous bird. Proceedings of the National Academy of Sciences. 2018;115(45):11549–54.

117. Threlfall CG, Williams NS, Hahs AK, Livesley SJ. Approaches to urban vegetation management and the impacts on urban bird and bat assemblages. Landscape and Urban Planning. 2016;153:28–39.

118. Le Roux DS, Ikin K, Lindenmayer DB, Manning AD, Gibbons P. The future of large old trees in urban landscapes. PLOS ONE. 2014;9(6):e99403. doi: 10.1371/journal.pone.0099403 24941258

119. Goddard MA, Dougill AJ, Benton TG. Why garden for wildlife? Social and ecological drivers, motivations and barriers for biodiversity management in residential landscapes. Ecological Economics. 2013;86:258–73.

120. Nassauer JI. Ecological function and the perception of suburban residential landscapes. Managing Urban and High Use Recreation Settings General Technical Report, USDA Forest Service North Central Forest Experiment Station, St Paul, MN. 1993;98–103.

121. Kiesling FM, Manning CM. How green is your thumb? Environmental gardening identity and ecological gardening practices. Journal of Environmental Psychology. 2010;30(3):315–27.

122. Beumer C. Show me your garden and I will tell you how sustainable you are: Dutch citizens’ perspectives on conserving biodiversity and promoting a sustainable urban living environment through domestic gardening. Urban Forestry & Urban Greening. 2018;30:260–79.

123. Helfand GE, Park JS, Nassauer JI, Kosek S. The economics of native plants in residential landscape designs. Landscape and Urban Planning. 2006;78(3):229–40.

124. McIntyre NE, Knowles-Yánez K, Hope D. Urban ecology as an interdisciplinary field: Differences in the use of “urban” between the social and natural sciences. Urban ecosystems. 2000;4(1):5–24.

125. De Winter JC. Using the student’s t-test with extremely small sample sizes. Practical Assessment, Research & Evaluation. 2013;18(10).

126. Ellis JI, Schneider DC. Evaluation of a gradient sampling design for environmental impact assessment. Environmental Monitoring and Assessment. 1997;48(2):157–72.

127. Telford RJ, Birks HJB. Effect of uneven sampling along an environmental gradient on transfer-function performance. Journal of Paleolimnology. 2011;46(1):99.

128. Lerman SB, Warren PS. The conservation value of residential yards: Linking birds and people. Ecological Applications. 2011;21(4):1327–39. 21774433

129. Larson KL, Casagrande D, Harlan SL, Yabiku ST. Residents’ yard choices and rationales in a desert city: Social priorities, ecological impacts, and decision tradeoffs. Environmental management. 2009;44(5):921. doi: 10.1007/s00267-009-9353-1 19777295

130. Harris EM, Polsky C, Larson KL, Garvoille R, Martin DG, Brumand J, et al. Heterogeneity in residential yard care: Evidence from Boston, Miami, and Phoenix. Human Ecology. 2012;40(5):735–49.

131. Yang J, Yan P, He R, Song X. Exploring land-use legacy effects on taxonomic and functional diversity of woody plants in a rapidly urbanizing landscape. Landscape and Urban Planning. 2017;162:92–103.

132. Uren HV, Dzidic PL, Bishop BJ. Exploring social and cultural norms to promote ecologically sensitive residential garden design. Landscape and Urban Planning. 2015;137:76–84.


Článek vyšel v časopise

PLOS One


2019 Číslo 9

Nejčtenější v tomto čísle

Tomuto tématu se dále věnují…