Adapting and applying evidence gathering techniques for planning and investment in street trees: A case study from Brisbane, Australia

Urban Forestry & Urban Greening 19 (2016) 79–87

Contents lists available at ScienceDirect

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

Adapting and applying evidence gathering techniques for planning and investment in street trees: A case study from Brisbane, Australia Lyndal Plant ∗ , Neil Sipe School of Geography, Planning and Environmental Management, University of Queensland, QLD 4072, Australia

a r t i c l e

i n f o

Article history: Received 24 August 2015 Received in revised form 6 July 2016 Accepted 8 July 2016 Available online 12 July 2016 Keywords: Pre-stratified sampling Street trees Urban forest structure and planning

a b s t r a c t Trees along footpath zones (or verges) grow on the “front-line” of urban forest ecosystems, increasingly recognised as essential to the quality of human life in cities. Growing so close to where residents live, work and travel, these street trees require careful planning and active management in order to balance their benefits against risks, liabilities, impacts and costs. Securing support and investment for urban trees is tough and robust business cases begin with accurate information about the resource. Few studies have accounted for spatial heterogeneity within a single land-use type in analyses of structure and composition of street tree populations. Remotely sensed footpath tree canopy cover data was used as a basis for stratification of random sampling across residential suburbs in the study area of Brisbane, Australia. Analysis of field survey data collected in 2010 from 80 representative sample sites in 52 suburbs revealed street tree population (432,445 ± 26,293) and stocking level (78%) estimates with low (6.08%) sampling error. Results also suggest that this population was transitioning to low risk, small-medium sized species with unproven longevity that could limit the capacity of the Brisbane’s Neighbourhood Shadeways planting program to expand from 35% footpath tree canopy cover in 2010, to a target of a 50% by 2031. This study advances the use of contemporary techniques for sampling extensive, unevenly distributed urban tree populations and the value of accurate resource knowledge to inform evidence-based planning and investment for urban forests. © 2016 Elsevier GmbH. All rights reserved.

1. Introduction Urban forest ecosystems provide beneficial services that are increasingly recognized as essential to the quality of life in human settlements, including a range of economic, environmental and social services that have been widely reported (Planet Ark, 2014; Dwyer et al., 2003; McPherson, 1995; McPherson et al., 1997, 2005; Nowak et al., 2006; Roy et al., 2012; Tarran, 2009). Ely and Pitman (2012) summarise the various benefits, acknowledged disservices (Dobbs et al., 2014; Lyytimäki and Sipilä, 2009), costs (McPherson and Peper, 1996) and risks of the street tree component of urban forests (Table 1). Although often a small subset of the urban forest, street trees within footpath zones (or verges), medians and other road reserve lands, grow close to where residents live, work, play and travel, requiring careful planning and active management in order to balance their benefits against the risks, liabilities, challenges and costs.

∗ Corresponding author at: School of Geography, Planning and Environmental Management, University of Queensland, Brisbane, St. Lucia, QLD 4072, Australia. E-mail addresses: [email protected] (L. Plant), [email protected] (N. Sipe). http://dx.doi.org/10.1016/j.ufug.2016.07.005 1618-8667/© 2016 Elsevier GmbH. All rights reserved.

In most cities, annual investment in street trees must also compete for limited local government funds with numerous other essential public assets, services, major projects and community priorities. Emerging urban forest research has increasingly focused on helping communities and urban tree managers to build a stronger evidence base to assist with strategic planning and to promote adequate investment (McPherson, 1995; Nowak et al., 2008a). The combination of “data-driven planning”, diverse funding sources, integrated within organisational priorities has also helped cities reorient tree planting towards broader green infrastructure goals (Young, 2011) including managing urban stormwater and reconnecting people with nature. Sustaining net benefits of urban forests over time requires the right kind of human intervention and management across three components − (i) the composition, condition and structure of the resource itself; (ii) a strong community framework and (iii) appropriate management of the resource (Clark et al., 1997; Kenney et al., 2011; Mincey et al., 2013). Miller et al. (2015) similarly suggests appropriate planning begins with urban forest managers asking typical asset management questions like “what they have”, “what they want to achieve” and “how to reach their goals”. The evidence base for planning and investment must be accurate and

80

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

Table 1 Environmental, economic and social benefits of street trees, adapted from Ely and Pitman (2012) and acknowledged disservices, costs and liabilities. Environmental benefits • Urban heat island effect mitigation • Wind speed modification • Carbon sequestration & storage • Avoided emissions (energy conservation) • Air quality improvement • Stormwater quality & quantity management • Soil stabilisation and nutrient recycling • Habitat support Social benefits • Human health and well being – physical, mental, and social capital • Cultural connections • Visual and aesthetic quality • Sense of place & time Economic benefits • Avoided costs of environmental regulatory and provisioning functions • Commercial vitality • Increased property values • Monetary values of human health and well-being benefits • Tourism • Preferred locations for corporate centres Disservices, liabilities and risks • Costs of planting, maintenance and removal/replacement • Damage to public infrastructure from tree roots, falling branches, canopy obstructions • Risks to public safety • Risks to adjacent private property • Storm damage and disruption to services • Obstruction of views and solar access • Nuisance and litter

relevant to target an audience of decision makers, community and broader stakeholders/potential investors yet at the same time align with contemporary conservation and urban forest management (Jansson and Lindgren, 2012; Sutherland et al., 2004; Wolf et al., 2015). Urban forest research, predominantly in the US, especially over the last three decades, has established quantifiable relationships between urban forest structure and ecosystem services functions and value. Tree canopy cover and stem density, species diversity, condition and distribution across urban landscapes not only affects the extent of ecosystem services, such as air and water cleaning and cooling services, but also determines the current and forecast levels of maintenance need, risk, resilience and capacity for enhancement (McPherson, 1995; Nowak et al., 1996). Software tools, also developed in the US, such as “i-Tree” (I-Tree, 2014), and tree canopy cover data from remotely sensed imagery are now available to assist cities around the world gather evidence for planning and managing urban trees. Several studies and sampling techniques have identified and accounted for spatial heterogeneity of urban tree canopy cover across different land-uses and tenures (Dobbs et al., 2013; Escobedo and Nowak, 2009; Jaenson et al., 1992; Kirkpatrick et al., 2011; Maco and McPherson, 2003; Nowak et al., 2008a; Nowak et al., 2008b; Sanders, 1984). Others have explored the influence of biophysical, land-use change and socio-economic factors on this uneven distribution and consequent inequity in urban ecosystem services provision (Conway and Bourne, 2013; Gong et al., 2013; Heynen et al., 2006; Ives and Kendal, 2014; Kendal et al., 2012; Landry and Chakraborty, 2009; Pham et al., 2013; Wolch et al., 2014). Such unevenness has not been limited to tree cover on private property, but is also found in public streetscapes and park-

Fig. 1. Location of Brisbane study area in South East Queensland, Australia.

lands. However, few studies and tools have accounted for spatial heterogeneity when sampling street tree populations within a single land-use type (Nagendra and Gopal, 2010). In large cities, 100% street tree inventories are often cost prohibitive or undertaken infrequently. Frequent monitoring of street tree assets, however, is especially important within areas of residential land use in rapidly growing cities where residential development can provide opportunities for improvements to public streetscapes as well as impacts on existing street trees. Such changes in street tree extent and structure, in turn, affect the flows of regulatory and cultural ecosystem services and disservices to the inhabitants of these populous land use zones (Berland and Hopton, 2014; Escobedo et al., 2015; McPherson et al., 2016; Sarkar et al., 2015; Tucker Lima et al., 2013; Dobbs et al., 2014). Sample surveys provide a cost effective alternative to monitor the street tree resource and inform forward planning (Nowak et al., 2008a). Given the importance of accurate evidence as the foundation to planning and investment in high value urban forest components like street trees, there are opportunities to improve contemporary evidence gathering techniques to ensure that sampling is representative of unevenly distributed tree cover. The aim of this study was to explore adaptations to evidence gathering techniques and demonstrate their usefulness for urban forest planning and policy review in the subtropical case study city of Brisbane, Australia. 1.1. Case study city: Brisbane, Australia In 2010, an estimated 1.06 million people were living within the local government area (LGA) of subtropical Brisbane, located, on the east coast of Australia at latitude 27◦ 28 S and longitude 153◦ 1 E (Fig. 1). Brisbane is the third most populated and one of the fastest growing cities in Australia (Australian Bureau of Statistics, 2011). Around 20,000 new residents each year, in the ten years 2001–2010 years, were attracted by employment opportunities, affordable housing choices, access to major health and transport facilities and other lifestyle features. Like most local councils in Australia, Brisbane City Council (BCC) has responsibility for the planning, planting, maintenance and protection of all trees on Council controlled land, including street trees. Brisbane’s challenge is to continue to strategically expand street and park tree cover, while maintaining and managing existing tree assets with limited resources. However, unlike other Australian capital cities, Brisbane’s area of jurisdiction extends well beyond the city centre to include 1340 square kilometres of residential, industrial, commercial centres, rural land uses and greenspace. To

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

Fig. 2. Percentage tree canopy cover by land tenure and land-use type, across overall Brisbane City Council (BCC) area and across residential suburbs of BCC in 2010.

facilitate cost effective street tree management and planning, accurate information about the extent, structure and composition of the street tree population is critical. 1.2. Brisbane urban forest overview In 2010, trees and their collective canopy areas covered 51.2% (70,673 ha) of the land area within the BCC boundaries. As shown in Fig. 2, almost half of Brisbane’s tree canopy cover was growing on public land, including just 4.1% (2960 ha) within road reserve lands along almost 4800 km of Brisbane streets. Within residential suburbs,1 where tree canopies covered 35.3% of the land area, most tree cover was on private land. Brisbane’s residential suburbs have generally grown a lot greener over the last 60 years (Plant, 1996a), including an estimated doubling of the proportion of properties with one or more adjacent street trees between 1961 and 2006 (Kirkpatrick et al., 2011). By 2010, the canopies of street trees growing within the footpath zone of road reserves, hereafter referred to as “footpath tree canopy cover”, covered an average of 35% of the total footpath zone area. Yet footpath tree canopy cover varied widely across residential suburbs (Fig. 3). Of Brisbane’s 115 residential suburbs only six had greater than 50% footpath tree canopy cover in 2010. Urban forest targets and strategies in Brisbane have continued to be set through a combination of BCC priorities, community and resource needs assessment. These include the first “no net canopy area loss” policy for tree replacement in an Australian city and community participatory initiatives, such as the “Neighbourhood Shadeways” program that aims to deliver attractive and shaded pathways for walking and cycling by increasing footpath tree canopy cover in residential areas from the 2010 average of 35% to 50% by 2031 (Brisbane City Council, 2013; Davison and Kirkpatrick, 2014; Plant, 2006). BCC has planted an average of around 11,000 street trees per year since 2007–08 (Brisbane City Council, 2013) including several species from the Sapindaceae family as well as species of the colourful and well adapted Handroanthus genus. Evergreen species that grow to a small to medium size, such as Harpullia pendula (tulipwood), Buckinghamia celcissima (ivory curl), Cupaniopsis anarcardioides (tuckeroo), and Xanthostemon chrysanthus (golden penda), have been preferred by the community and most commonly used since Brisbane’s Neighbourhood Shadeway pro-

1 Residential suburbs are defined as those suburbs with 50% or more of their land area designated in Brisbane City Council – City Plan 2000 as a type of residential zoning.

81

gram began in 2006–07 (Brisbane City Council, 2013). In 2009–10 almost a fifth of all street trees planted were H. pendula. While other Australian cities like Sydney, with 15.5% tree cover in 2008, Melbourne with 11% tree cover in 2013 (City of Melbourne, 2013; City of Sydney, 2013), are striving to increase tree canopy cover over time as part of their plans to sustain cleaner, greener and healthier habitats for urban residents. Brisbane’s challenge may be maintaining their 35.3% canopy cover across residential suburbs. Significant changes in Brisbane’s tree cover are predicted as patterns of residential development change to accommodate increasing population (Brisbane City Council, 2014). Privately owned houses with backyards, that have dominated residential land use, provided the space for almost two thirds of residential tree cover in 2010. This is changing, especially in high growth Brisbane suburbs as new dwelling forms occupy a greater proportion of site area (Daniel et al., 2016; Hall, 2010). Streetscapes, in cities like Brisbane, are likely to become important components of diverse greenspace networks needed to meet the demands of densifying cities and their changing climates (Byrne et al., 2010; Hamin and Gurran, 2009). The extent to which tree cover in public spaces like parks and streets can help compensate for losses on private land may depend on significant and strategic investments in urban greening. 2. Data and methods We used data from spatial imagery analysis to pre-stratify random sampling to be representative of the unevenness of footpath tree canopy cover and density in residential areas, and then used the sample survey data to explore the extent, structure, composition and management needs of street tree assets in the case study city. Each of the four steps taken in the methods – (i) pre-stratification, (ii) sampling and (iii) street tree survey and (iv) survey analysis – are described below. 2.1. Pre-stratification 2010 tree cover data was obtained from BCC and used to prestratify the sampling of the street tree population across residential areas of Brisbane. BCC had developed tree cover data at two metre resolution, using two types of spatial data. Overstorey foliage projective cover (FPC) (Armston et al., 2009), was derived from airborne laser scanner data (LiDAR), acquired by the state government Department of Environment and Resource Management (DERM) in 2009. FPC data was then combined with analysis of multi-spectral WorldView 2 satellite imagery, acquired in 2010 from Digital Globe to fine tune and update FPC (Arroyo et al., 2010). In our study, BCC tree cover data provided a base layer of spatial information which was intersected with the polygon vector layer for footpath zone parcels in residential suburbs, and analysed using a desktop geographic information system (ArcGIS v10.0), to derive a separate footpath tree canopy cover spatial layer. Two steps were used to identify strata which were representative of the uneven distribution of footpath tree coverage across the 115 residential suburb boundaries of the case study area. First, 500 × 500 m grid cells were overlayed on the footpath tree canopy cover spatial layer. Second, to account for variations in the density of street trees within footpath zones, not revealed by canopy coverage metrics alone, a surrogate measure termed “street tree canopy clustering” was defined as the average tree canopy clump (an area of contiguous tree canopy) size within the footpath zone, multiplied by the number of clumps within each grid cell. The range and frequency of street tree canopy clustering within the grid cells is shown in Fig. 4. The spatial distribution of the ten percentile strata capturing this variation across Brisbane’s residential suburbs is shown in Fig. 5.

82

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

Fig. 3. Variation in street tree canopy cover levels across BCC’s 188 mainland suburbs in 2010 (BCC map).Suburbs in Brisbane are geographical boundaries defined by the Australian Geographical Standard Classification (ASGC 2001) (ABS 2011).

the second stage stratification suggested by Jaenson et al. (1992). However, remote sensing data analysis rather than ground based pre-sampling was used to pre-stratify by street tree density and coverage (Maco and McPherson, 2003). 2.3. Street tree survey

Fig. 4. Frequency distribution of street tree canopy clustering (average canopy clump area in square metres X number of clumps) across 500 × 500 m grid cells in residential suburbs of Brisbane in 2010. Ten percentiles of the distribution are marked with dashed lines.

2.2. Sampling Eight grid cell sites were chosen randomly from each of the ten percentile strata to become the 80 sites for on-ground street tree sampling. This differs from the approach used in other studies that stratify the sampling of the street tree population by land-use or governance boundaries (I-Tree 2014) and then select sample sites randomly within those strata. Using land-use or other boundaries to stratify sampling can allow output comparisons to be made across strata. However if heterogeneity within strata is not accounted for in the sampling, then such comparisons become less accurate. This study used a sampling approach similar to

Each sample site was used as the starting point for the 2010 on-ground street tree inventory survey. Along no less than two kilometres of street length, street trees and planting opportunities in footpath zones on both sides of the street were mapped and attributes of all located street trees recorded. Inventory data collected by BCC contractors included: length of street(s) surveyed, location of trees, presence or absence of overhead powerlines, number of trees, number of planting opportunities,2 tree species, age (new: 0–2 years, juvenile: 3–5 years, maturing: 6–15 years, mature: 16–30 years, aged: >30 years), tree size (diameter at breast height – DBH, and canopy width), tree condition (dead, poor, fair, good, excellent) and tree risk rating and tree maintenance needs. Survey contractors were qualified arborists and used a BCC tree risk assessment standard adapted from AS/NZS ISO 31000:2009 Risk Management standard (Standards Australia, 2009) to rate tree risk as low medium or high. Maintenance needs were categorised as none at this time, minor prune (work that can be undertaken from the ground level), medium prune (work that can be undertaken from a 55 ft elevated platform vehicle), major work (requiring specialised equipment), and removal recommended (for trees iden-

2 Planting opportunities were determined using BCC streetscape design guidelines for street tree planting locations in City Plan 2000 http://www.brisbane.qld. gov.au/planning-building/planning-guidelines-and-tools/superseded-brisbanecity-plan-2000/centres-design-detail-manual/streetscape-design.

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

83

Fig. 5. Spatial distribution of the ten percentile categories of street tree canopy clustering across the residential BCC area.

tified as high risk or dead). Data was collected for 16,669 street trees, from 80 sample sites, along 220 km (or 4.5%) of the 4839 km of residential street/road length in residential Brisbane.

Rowntree, 1989; Miller et al., 2015; Peper et al., 2007; Santamour, 1990), population profile and management priorities were identified for street trees in Brisbane.

2.4. Survey analysis

3. Results

The survey data was extrapolated to provide an estimate of residential street tree population and planting opportunities using the average number of trees per kilometre from each of the strata multiplied by the total length of streets of each strata. Standard errors were calculated from the deviations from the mean number of trees per kilometre per strata. Total relative standard error was the sum of strata standard errors as a proportion of tree counts per strata. Using comparisons to both best practice guidelines, ShannonWeiner diversity (Shannon, 1948) and evenness indices and structural analyses of street tree populations of other tropical,3 subtropical, Australian and international cities (City of Melbourne, 2013; City of Sydney, 2013; Dobbs et al., 2013; McPherson and

3.1. Street tree population, stocking level and sampling error The estimated number of street trees growing along residential Brisbane footpaths in 2010 was 432,445 (±26,293). The 16,669 street trees of the sample represented just 3.6–4.1% of the estimated street tree population yet total relative standard error for the sampling was limited to 6.08%. There was an average of 89 street trees per kilometre, and a range of 69–113 street trees per kilometre. There were an estimated 123,222 (±7492) planting opportunities, meaning that street trees were already occupying around 78% of available spaces in Brisbane. 3.2. Street tree composition and diversity

3 Evenness is a measure of how evenly individuals are distributed among different species. The Shannon-Weiner (H’) index of diversity accounts for both abundance and evenness of the species in a population and calculated as the natural log of the proportion of each species relative to the total number of species is summed across species and multiplied by −1.

Over 200 different species of trees were found growing in Brisbane footpath zones in the sample area. This abundance of species is reflected in the high Shannon-Weiner diversity index of 4.1. However just 30 species make up 73% of population, and ten species

84

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

Fig. 6. Ten most common species of the 2010 Brisbane street tree sample population (% sample popn) and their contributions in% canopy area.

Table 2 Species diversity of 2010 street tree population in Brisbane, Melbourne (2008) and Sydney (2011) compared to Santamour (1990), “30:20:10” guideline.

Brisbanea Melbourneb Sydneyb Guideline a b

% of same family

% of same genus

% of same species

34.5 43 40 30

8 24 N/A 20

8.5 12 9.5 10

Street trees. Street and park trees.

made up almost 50% of the population. This uneven species distribution was reflected in an evenness level of 0.77. Caesalpinia ferrea (leopard tree), planted mostly in the 1970s and 1980s, were found to be not just the most common species (8.9%), but also the species providing the greatest canopy area contribution (Fig. 6). C. ferrea was the only non-native deciduous species in the top ten − the majority of others were native evergreen species such as H. pendula, and B. celcissima. Aside from two larger growing tree species, Flindersia australis (crows Ash) and Jacaranda mimosaefolia (jacaranda), the most common species found on Brisbane footpaths in the sample survey were species that grow to a small-medium size. 33% of Brisbane’s most common street tree species were from the Myrtaceae family, which is susceptible to Myrtle Rust. As shown in Table 2, this was just over the level of 30% of species from the same family (no more than 20% of species from the same genus, and no more than 10% of species from the same species) suggested as a diversity guideline by Santamour (1990), and much less than other Australian cities of Melbourne and Sydney dominance of myrtaceous trees.

Fig. 7. Street tree age classes identified in the 2010 Brisbane street tree sample. Table 3 Size range (in mm Diameter at Breast Height-DBH), of the ten (10) most common street tree species in the sample. DBH (mm)

0–150

151–300

301–450

451–600

>600

Caesalpinea ferra Harpullia pendula Buckinghamia celcissima Delonix regia Syzygium leuhmanii Lophostemon confertus Xanthostemon chrysanthus Cupaniopsis anacardioides Flindersia australis Jacaranda mimosaefolia TOTAL (all species)

44 70 78 20 67 82 86 66 85 51 57

41 26 17 44 32 13 14 32 14 34 31

12 3 5 22 1 4 0 2 1 10 8

3 0 0 9 0 1 0 0 0 5 3

1 0 0 5 0 0 0 0 0 0 1

3.3. Street tree age, size, health, risk and maintenance needs profile Most of Brisbane’s street trees have not yet reached maturity, with more than 70% estimated to be 15 years old or less (Fig. 7). Only 2% were categorized as “aged”. 42% of the sampled trees were no taller than 5 m and almost a third of street trees in the sample survey were growing under overhead powerlines. Average tree canopy area across the sample was 21.87 m2 . Of the top ten most common street trees in the sample, only three have more than 10% of their population greater than 300 mm (or 6inches) in trunk DBH (Table. 3). Most street trees were rated in fair health in the 2010 sample survey. The most common species, C. ferrea, were mostly of fair

Fig. 8. Street tree maintenance needs categories identified in 2010 Brisbane street tree sample survey. Std EPV = work that can be performed from a 55 ft Elevated Platform Vehicle.

health irrespective of age category. Lophostemon confertus (brush box) was found to be the most common species of maturing age in good health. Those in poor health or dead, made up the 4.4% of the street tree sample population. 99% of the street trees were rated as low risk. in the 2010 street tree survey and 75.4% of street tree maintenance work identified in the 2010 survey was non-urgent minor pruning (Fig. 8), consistent

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

with a low risk, fair health, maturing population. In summary, the street tree survey results revealed an extensive population of small to medium sized, diverse, maturing, low risk street trees of fair health.

4. Discussion Pre-stratifying to account for spatial heterogeneity in street tree density and canopy cover significantly reduced forecast sampling error. Small sample sizes often limit the validity of their extrapolations. Nowak et al. (2008b) suggested sample sizes for assessing urban forest structure, of between 50 and 100, could attract sampling error of between 25 and 15%. i-Tree Streets (2014) recommends sampling 3–6% of a city’s street segments to achieve around 10% standard error. The approach used in this study reduced relative sampling error to 6.08% using just 80 sample site street segments, representing 3.6–4.1% of the estimated street tree population and 4.5% of residential road length, providing a robust population estimate and profile of a vast street tree resource in Brisbane. Remotely sensed data application to urban forest management continues to expand (Zhang and Qiu, 2012). Combinations of satellite and airborne LiDAR imagery, as sourced for this study, are also more accessible to urban tree managers as local government authorities begin to use such data to develop detailed land cover updates (Norton et al., 2015). Access to remotely sensed data not only allows more evidence about urban forest assets to be sourced at the “desk-top”, but provides the basis for accurate and efficient field survey data collection, especially in large scale study areas where uneven urban tree density and distribution is the norm. In this case, “desk-top” street tree canopy cover data replaced the presample “drive-by” survey of street tree density and street segment length estimation recommended for street tree sample surveys (Jaenson et al., 1992; Maco and McPherson, 2003). When applied across residential suburbs in Brisbane, analysis of a representative sample of the street trees revealed an extensive tree population transitioning away from larger growing species to small and medium size, diverse types of maturing, low risk trees. This type of evidence is fundamental to guiding future planning and investment, including monitoring progress towards resource management goals and ecosystem service goals such as shading and cooling. The mild climate of Brisbane supported a slightly greater level of species diversity (H’ = 4.1) than the street tree population of the wet temperate Australian city of Melbourne (H’ = 3.7) (Dobbs et al., 2013) but much greater than those reported in tropical street tree populations (Jim and Liu, 2001; Pedlowski et al., 2002; Sreetheran et al., 2011; Thaiutsa et al., 2008; Deb et al., 2013). Although climate can influence species richness (Cowett and Bassuk, 2014; McPherson and Rowntree, 1989), Brisbane’s diversity of street tree species may also be a product of influential visionaries (Plant, 1996b), pre-development land use and planting policies (Kendal et al., 2012). Contemporary local government decision-makers can similarly help sustain healthy and beneficial street tree assets over time by continuing to support a mix of tree species, sizes and ages (McPherson et al., 2005). While not as vulnerable as Sydney and Melbourne to losses from the threat of Myrtle Rust, our analysis confirmed the trend towards a range of native, non-Myrtaceae, small-medium evergreen species such as B. celcissima (Family Proteacea),was occurring at the expense of some larger growing exotic species like C. ferrea (leopard tree) and D. regia (poinciana). More than a third of Brisbane’s street tree population in 2010 were small and medium growing species. Downsizing of urban trees is often a symptom of above and below ground space constraints, and a common

85

response to the issues of infrastructure damage and risk (Ely, 2010; McPherson et al., 2005). Although small-medium growing evergreen species align well with risk management goals and powerline constraints of Brisbane’s streetscapes, there are trade-offs in ecosystem services to downsizing street tree species (McPherson and Rowntree, 1989). City of Sydney is forecasting that 57% of their street trees will grow large enough to tackle urban heat island effects in low tree cover suburbs (City of Sydney, 2013). Brisbane’s street trees surveyed in 2010 were on average just over 5 m tall. In 2007, New York City’s (NYC) street tree population was around the same in numbers as Brisbane’s, yet the majority of NYC street trees were greater than 10 m tall. NYC’s large shady trees were also delivering valuable environmental benefits, especially cooling urban heat islands, energy conservation and air quality improvement (Peper et al., 2007). Greener, shaded and cooler streetscapes, especially in subtropical climates, can also support healthy, active living as residents increase the frequency and duration of their neighbourhood walks (Heart Foundation, 2014; Sarkar et al., 2015; Nagendra and Gopal, 2010). Although we found space available for an estimated 123,000 additional street trees, BCC’s goal of increasing residential footpath tree canopy cover from 35% to 50% by 2031 may be challenged by downsizing of the tree species. Five of the small-medium transition species were amongst the top ten of the surveyed street tree population, however, they were contributing just 11.47% in canopy area compared to the 23% of just two of the larger growing species (C. ferrea and D. regia).To attain a 15% increase in footpath tree canopy area from the maximum potential residential street tree population in Brisbane would require the average individual footpath tree canopy area to be more than twice that of the sample surveyed (ie. 45 m2 compared to the sample average of 21.87 m2 ) by 2031. BCC Streetscape Design Guidelines (Brisbane City Council, 2014) promote a mixture of species, including larger growing and iconic tree species, supported by a range of tree planting pit and trench designs, to create sub-tropical boulevards along major roads. Our study results suggest the need for a similar approach within residential streets, by using a diversity of at least medium growing tree species, at closer spacings, from some proven families such as Sapindaceae, as alternatives to small species like B.celsissima and vulnerable Myrtaceae species like X. chrysanthus. The tolerance of residents to mixtures of medium sized street tree species within the streetscape deserves further investigation. Meanwhile, implementation of policies such as a “no net canopy area loss” approach to tree replacement (Brisbane City Council, 2014) would also need to be maintained to progress towards the 50% footpath tree canopy cover target. Perhaps as a consequence of the high proportion of smallmedium sized trees, this study identified very low levels of risk to people or property from the Brisbane street tree population. Yet there need only be perceptions of higher risk from street trees by the community (Kirkpatrick et al., 2012), to demand long-lasting and ongoing trade-offs in species size and stocking levels. What may be gained in lowering risks from such trade-offs may not only reduce environmental benefits but cost local government more in other ways. Small-medium growing species are more likely to require ongoing minor pruning to clear branches from footpath zones for pedestrians. This, together with a population dominated by street trees of maturing age, may already be contributing to the high proportion of ground based minor pruning maintenance needs identified in Brisbane. Unlike larger growing species which respond to earlier formative pruning to grow clear of footpath and road clearance zones, smaller growing species never “grow out” of the need for regular ground based clearance pruning. Rather than reducing ongoing costs of maintenance, smaller growing, less

86

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87

long-lived street tree species, can increase tree management budget requirements and reduce benefits (McPherson and Rowntree, 1989). With a small proportion of aged trees (2%), and a generally uneven age profile, Brisbane’s street tree population has very low vulnerability to large losses of aged trees. However the large proportion (66%) of street trees of a maturing age class may indicate two possible scenarios. First, some bias in the survey results from difficulties in assessing the age of smaller growing species. Second, a large proportion of maturing trees could be a symptom of premature decline and losses that deserve further investigation using survival rate surveys of samples of 3, 5 and 10 year old plantings. Unlike species such as D. regia and C. ferrea that have proven their resilience by reaching larger size and older age, newer species have not yet proven their performance over time. Brisbane’s tree management currently includes a program of establishment and juvenile maintenance for street trees up to age five (Brisbane City Council, 2013). “Maturing” street trees, however, receive predominantly reactive maintenance. One or two additional programmed maintenance visits for “maturing” street trees may improve longevity, offer the opportunity to gather survival rate and growth rate data and control costs in the longer term. Both hazard reduction and formative structural work, while the trees are still small enough to work on from the ground, has been recognised as a key component of proactive cost and tree risk management (Miller et al., 2015). 5. Conclusion The extent of tree cover, in particular the composition and structure of street tree populations, is increasingly informed by the analysis of the combination of remotely sensed information and sample field inventories. Accounting for the typically uneven distribution and density of street trees is important for efficient and accurate sample surveys. Our “desk-top” pre-stratification using a combination of street tree stem density and coverage provided a more representative basis for sample surveying. In the case study city, survey data was analysed to help review policies and progress towards footpath tree cover targets and recommend priorities for future planning and investment in Brisbane’s street trees. Those cities that progress best towards sustainable, multi-purpose, valued urban forests, are those that have combined several accurate sources of evidence to review and redirect their policies, investment and actions. Adapting evidence gathering methods, such as the use of remotely sensed tree canopy attributes to stratify field sampling, provides the opportunity for less extensive and more frequent monitoring of typically dynamic and heterogeneous urban forest components in other cities. Acknowledgements We thank two anonymous reviewers for comments that helped improve the manuscript. We also thank the Brisbane City Council, in particular, Mathew Byrne and Amanda Smedts for assistance with the development of the stratification for the sample survey and Jason Beale for sharing both spatial and street tree sample survey data, under licence agreement. This research was funded by an Australian Postgraduate Award. References Armston, J.D., Denham, R.J., Danaher, T.J., Scarth, P.F., Moffiet, T.N., 2009. Prediction and Validation of Foliage Projective Cover from Landsat-5 TM and Landsat-7 ETM+ Imagery| NOVA. The University of Newcastle’s Digital Repository, International Society for Optical Engineering (SPIE). Arroyo, L.A., Johansen, K., Armston, J., Phinn, S., 2010. Integration of LiDAR and QuickBird imagery for mapping riparian biophysical parameters and land

cover types in Australian tropical savannas. For. Ecol. Manag. 259, 598–606, http://dx.doi.org/10.1016/j.foreco.2009.11.018. Australian Bureau of Statistics, 2011. Census Quick Stats, Brisbane – Dwellings [Online], Available: http://www.censusdata.abs.gov.au/census services/ getproduct/census/2011/quickstat/3001?opendocument&navpos=220 (accessed 9.06.14). Berland, A., Hopton, M.E., 2014. Comparing street tree assemblages and associated stormwater benefits among communities in metropolitan Cincinnati, Ohio, USA. Urban For. Urban Green. 13, 734–741, http://dx.doi.org/10.1016/j.ufug. 2014.06.004. Brisbane City Council, 2013. Street Trees [Online], Available: http://www.brisbane. qld.gov.au/environment-waste/natural-environment/plants-trees-gardens/ brisbanes-trees/street-trees/index.htm (accessed 30.06.13). Brisbane City Council, 2014. City Plan 2014 [Online], Available: http://www. brisbane.qld.gov.au/planning-building/planning-guidelines-tools/brisbanecity-plan-2014 (accessed 28.03.15). Byrne, J., Sipe, N., Searle, G., 2010. Green around the gills? The challenge of density for urban greenspace planning in SEQ. Aust. Plann. 47, 162–177, http://dx.doi. org/10.1080/07293682.2010.508204. City of Melbourne, 2013. Urban Forest Strategy [Online], Available: http://www. melbourne.vic.gov.au/Sustainability/UrbanForest/Pages/About.aspx (accessed 17.06.13). City of Sydney, 2013. Urban Forest Strategy [Online], Available: http://www. cityofsydney.nsw.gov.au/ data/assets/pdf file/0003/132249/Urban-ForestStrategy-Adopted-Feb-2013.pdf (accessed 26.08.13). Clark, J.R., Matheny, N.P., Cross, G., Wake, V., 1997. A model of urban forest sustainability. J. Arboriculture 23, 17–30. Conway, T.M., Bourne, K.S., 2013. A comparison of neighborhood characteristics related to canopy cover, stem density and species richness in an urban forest. Landsc. Urban Plann. 113, 10–18, http://dx.doi.org/10.1016/j.landurbplan. 2013.01.005. Cowett, F.D., Bassuk, N.L., 2014. Statewide assessment of street trees in New York State, USA. Urban For. Urban Green. 13, 213–220, http://dx.doi.org/10.1016/j. ufug.2014.02.001. Daniel, C., Morrison, T.H., Phinn, S., 2016. The governance of private residential land in cities and spatial effects on tree cover. Environ. Sci. Policy 62, 79–89. Davison, A., Kirkpatrick, J., 2014. Re-inventing the urban forest: the rise of arboriculture in Australia. Urban Policy Res. 32, 145–162, http://dx.doi.org/10. 1080/08111146.2013.832669. Deb, J.C., Halim, M.A., Rahman, H.M.T., Al-Ahmed, R., 2013. Density, diversity, composition and distribution of street trees in Sylhet Metropolitan City of Bangladesh. Arboricultural J. 35, 36–49, http://dx.doi.org/10.1080/03071375. 2013.770656. Dobbs, C., Kendal, D., Nitschke, C., 2013. The effects of land tenure and land use on the urban forest structure and composition of Melbourne. Urban For. Urban Green. 12, 417–425, http://dx.doi.org/10.1016/j.ufug.2013.06.006. Dobbs, C., Kendal, D., Nitschke, C.R., 2014. Multiple ecosystem services and disservices of the urban forest establishing their connections with landscape structure and sociodemographics. Ecol. Indic. 43, 44–55, http://dx.doi.org/10. 1016/j.ecolind.2014.02.007. Dwyer, J.F., Nowak, D.J., Noble, M.H., 2003. Sustaining urban forests. J. Arboriculture 29, 49–55, http://dx.doi.org/10.1016/j.ufug.2006.01.007. Ely, M.E., 2010. Integrating Trees into the Design of the City: Expert Opinions on Developing More Sustainable Practices for Planting Street Trees in Australian Cities. PhD Thesis. The University of Adelaide. Escobedo, F.J., Nowak, D.J., 2009. Spatial heterogeneity and air pollution removal by an urban forest. Landsc. Urban Plann. 90, 102–110, http://dx.doi.org/10. 1016/j.landurbplan.2008.10.021. Escobedo, F.J., Adams, D.C., Timilsina, N., 2015. Urban forest structure effects on property value. Ecosyst. Serv. 12, 209–217, http://dx.doi.org/10.1016/j.ecoser. 2014.05.002. Gong, C., Yu, S., Joesting, H., Chen, J., 2013. Determining socioeconomic drivers of urban forest fragmentation with historical remote sensing images. Landsc. Urban Plann. 117, 57–65, http://dx.doi.org/10.1016/j.landurbplan.2013.04.009. Hall, T., 2010. Life and Death of the Australian Backyard. CSIRO Publishing, Collingwood, Victoria. Hamin, E.M., Gurran, N., 2009. Urban form and climate change: balancing adaptation and mitigation in the US and Australia. Habitat Int. 33, 238–245. Heart Foundation, 2014. Blueprint for an Active Australia [Online], Available: http://heartfoundation.org.au/for-professionals/built-environment (accessed 18.10.15). Heynen, N., Perkins, H.A., Roy, P., 2006. The political ecology of uneven urban green space the impact of political economy on race and ethnicity in producing environmental inequality in Milwaukee. Urban Aff. Rev. 42, 3–25. I-Tree, 2014. Tools for Assessing and Managing Community Forests [Online], Available: http://www.itreetools.org/ (accessed 27.04.14). Ives, C.D., Kendal, D., 2014. The role of social values in the management of ecological systems. J. Environ. Manag. 144, 67–72, http://dx.doi.org/10.1016/j. jenvman.2014.05.013. Jaenson, R., Bassuk, N., Schwager, S., Headley, D., 1992. A statistical method for the accurate and rapid sampling of urban street tree populations. J. Arboriculture 18, 171–183. Jansson, M., Lindgren, T., 2012. A review of the concept ‘management’ in relation to urban landscapes and green spaces: toward a holistic understanding. Urban For. Urban Green. 11, 139–145, http://dx.doi.org/10.1016/j.ufug.2012.01.004.

L. Plant, N. Sipe / Urban Forestry & Urban Greening 19 (2016) 79–87 Jim, C.Y., Liu, H.T., 2001. Species diversity of three major urban forest types in Guangzhou City, China. For. Ecol. Manag. 146, 99–114, http://dx.doi.org/10. 1016/S0378-1127(00)00449-7. Kendal, D., Williams, N.S.G., Williams, K.J.H., 2012. Drivers of diversity and tree cover in gardens, parks and streetscapes in an Australian city. Urban For. Urban Green. 11, 257–265, http://dx.doi.org/10.1016/j.ufug.2012.03.005. Kenney, W.A., Van Wassenaer, P.J., Satel, A.L., 2011. Criteria and indicators for strategic urban forest planning and management. Arboriculture Urban For. 37, 108–117. Kirkpatrick, J.B., Daniels, G.D., Davison, A., 2011. Temporal and spatial variation in garden and street trees in six eastern Australian cities. Landsc. Urban Plann. 101, 244–252, http://dx.doi.org/10.1016/j.landurbplan.2011.02.029. Kirkpatrick, J.B., Davison, A., Daniels, G.D., 2012. Resident attitudes towards trees influence the planting and removal of different types of trees in eastern Australian cities. Landsc. Urban Plann. 107 (2), 147–158, http://dx.doi.org/10. 1016/j.landurbplan.2012.05.015. Landry, S.M., Chakraborty, J., 2009. Street trees and equity: evaluating the spatial distribution of an urban amenity. Environ. Plann. A 41, 2651. Lyytimäki, J., Sipilä, M., 2009. Hopping on one leg – The challenge of ecosystem disservices for urban green management. Urban For. Urban Green. 8, 309–315, http://dx.doi.org/10.1016/j.ufug.2009.09.003. Maco, S.E., McPherson, E.G., 2003. A practical approach to assessing structure, function, and value of street tree populations in small communities. J. Arboriculture 29, 84–97. McPherson, E.G., 1995. Net benefits of healthy and productive urban forests. In: Bradley, G.A. (Ed.), Urban Forest Landscapes, Integrating Multidisciplinary Perspectives. University of Washington Press, Seattle. McPherson, E.G., Nowak, D., Heisler, G., Grimmond, S., Souch, C., Grant, R., Rowntree, R., 1997. Quantifying urban forest structure, function, and value: the Chicago Urban Forest Climate Project. Urban Ecosyst. 1, 49–61. McPherson, E.G., Peper, P.P., 1996. Costs of street tree damage to infrastructure. Arboricultural J. 20 (2), 143–159 http://www.tandfonline.com/doi/abs/10. 1080/03071375.1996.9747110. McPherson, E.G., Rowntree, R.A., 1989. Using structural measures to compare twenty-two U.S. street tree populations. Landsc. J. 8, 13–23, http://dx.doi.org/ 10.3368/lj.8.1.13. McPherson, E.G., Van Doorn, N., De Goede, J., 2016. Structure, function and value of street trees in California, USA. Urban For. Urban Green. 17, 104–115, http://dx. doi.org/10.1016/j.ufug.2016.03.013. McPherson, G., Simpson, J.R., Peper, P.J., Maco, S.E., Xiao, Q., 2005. Municipal forest benefits and costs in five US cities. J. For. 103, 411–416. Miller, R.W., Hauer, R.J., Werner, L.P., 2015. Urban Forestry: Planning and Managing Urban Greenspaces. Waveland Press. Mincey, S., Hutten, M., Fischer, B., Evans, T., Stewart, S., Vogt, J., 2013. Structuring institutional analysis for urban ecosystems: a key to sustainable urban forest management. Urban Ecosyst. 16, 553–571, http://dx.doi.org/10.1007/s11252013-0286-3. Nagendra, H., Gopal, D., 2010. Street trees in Bangalore: density, diversity, composition and distribution. Urban For. Urban Green. 9, 129–137, http://dx. doi.org/10.1016/j.ufug.2009.12.005. Norton, B.A., Coutts, A.M., Livesley, S.J., Harris, R.J., Hunter, A.M., Williams, N.S.G., 2015. Planning for cooler cities: a framework to prioritise green infrastructure to mitigate high temperatures in urban landscapes. Landscape Urban Plan. 134, 127–138 http://dx.doi.org/10.1016/j.landurbplan.2014.10.018. Nowak, D.J., Rowntree, R.A., McPherson, E.G., Sisinni, S.M., Kerkmann, E.R., Stevens, J.C., 1996. Measuring and analyzing urban tree cover. Landsc. Urban Plann. 36, 49–57. Nowak, D.J., Crane, D.E., Stevens, J.C., 2006. Air pollution removal by urban trees and shrubs in the United States. Urban For. Urban Green. 4, 115–123. Nowak, D.J., Crane, D.E., Stevens, J.C., Hoehn, R.E., Walton, J.T., Bond, J., 2008a. A ground-based method of assessing urban forest structure and ecosystem services. Arboriculture Urban For. 34, 347–358. Nowak, D.J., Walton, J.T., Stevens, J.C., Crane, D.E., Hoehn, R.E., 2008b. Effect of plot and sample size on timing and precision of urban forest assessments. Arboriculture Urban For. 34, 386–390.

87

Pedlowski, M.A., Da Silva, V.a.C., Adell, J.J.C., Heynen, N.C., 2002. Urban forest and environmental inequality in Campos dos Goytacazes, Rio de Janeiro, Brazil. Urban Ecosyst. 6, 9–20, http://dx.doi.org/10.1023/A:1025910528583. Peper, P.J., McPherson, E.G., Simpson, J.R., Gardner, S.L., Vargas, K.E., Xiao, Q., Watt, F., 2007. New York City, New York: Municipal Forest Resource Analysis. Center for Urban Forest Research, USDA Forest Service, Pacific Southwest Research Station. Pham, T.-T.-H., Apparicio, P., Landry, S., Séguin, A.-M., Gagnon, M., 2013. Predictors of the distribution of street and backyard vegetation in Montreal, Canada. Urban For. Urban Green. 12, 18–27, http://dx.doi.org/10.1016/j.ufug.2012.09. 002. Planet Ark, 2014. Valuing Trees: What Is Nature Worth? [Online], Available: http:// treeday.planetark.org/documents/doc-1170-valuing-trees-report?-full-final. pdf (accessed 23.06.15). Plant, L., 1996a. Brisbane’s urban forest: past influences, current trends and future needs. Trees Nat. Resour. 38, 6. Plant, L., 1996b. Brisbane’s urban forests. Past influence, current trends and future need. Trees Nat. Resour. 30, 6. Plant, L., 2006. Brisbane: beautiful one day, perfect the next-Is there room for improvement. In: Lawry, D., Christophel, D., Smith, S. (Eds.), Proceedings of 7th National Street Tree Symposium. University of Adelaide, Australia, pp. 89–94. 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 Green. 11, 351–363, http://dx.doi.org/10.1016/j.ufug. 2012.06.006. Sanders, R.A., 1984. Some determinants of urban forest structure. Urban Ecol. 8, 13–27. Santamour Jr., R.S., 1990. Trees and urban planting: diversity, uniformity and common sense. In: Proceedings of the 7th Conference of the Metropolitan Tree Improvement Alliance (METRIA) Conference, Lisle, IL, pp. 57–65. Sarkar, C., Webster, C., Pryor, M., Tang, D., Melbourne, S., Zhang, X., Jianzheng, L., 2015. Exploring associations between urban green, street design and walking: results from the Greater London boroughs. Landsc. Urban Plann. 143, 112–125, http://dx.doi.org/10.1016/j.landurbplan.2015.06.013. Shannon, C.E., 1948. The mathematical theory of communication. Bell Syst. Tech. J., 3–91. Sreetheran, M., Adnan, M., Khairil Azuar, A., 2011. Street tree inventory and tree risk assessment of selected major roads in Kuala Lumpur, Malaysia. Arboriculture Urban For. 37, 226. Standards Australia, 2009. AS/NZS ISO 31000:2009. Risk Management — Principles and Guidelines (Online). SAI Global Limited (accessed 18.07.15). Sutherland, W.J., Pullin, A.S., Dolman, P.M., Knight, T.M., 2004. The need for evidence-based conservation. Trends Ecol. Evol. 19, 305–308, http://dx.doi.org/ 10.1016/j.tree.2004.03.018. Tarran, J., 2009. People and trees: providing benefits, overcoming impediments. In: Lawry Oam, D., Gardner, J., Merrett, B. (Eds.), Proceedings of the 10th National Street Tree Symposium. Adelaide, Australia. Adelaide University, pp. 63–82. Thaiutsa, B., Puangchit, L., Kjelgren, R., Arunpraparut, W., 2008. Urban green space, street tree and heritage large tree assessment in Bangkok, Thailand. Urban For. Urban Green. 7, 219–229, http://dx.doi.org/10.1016/j.ufug.2008.03.002. Tucker Lima, J.M., Staudhammer, C.L., Brandeis, T.J., Escobedo, F.J., Zipperer, W., 2013. Temporal dynamics of a subtropical urban forest in San Juan, Puerto Rico, 2001–2010. Landsc. Urban Plann. 120, 96–106, http://dx.doi.org/10.1016/ j.landurbplan.2013.08.007. Wolch, J.R., Byrne, J., Newell, J.P., 2014. Urban green space, public health, and environmental justice: the challenge of making cities ‘just green enough’. Landsc. Urban Plann. 125, 234–244, http://dx.doi.org/10.1016/j.landurbplan. 2014.01.017. Wolf, K.L., Measells, M.K., Grado, S.C., Robbins, A.S.T., 2015. Economic values of metro nature health benefits: a life course approach. Urban For. Urban Green. 14, 694–701, http://dx.doi.org/10.1016/j.ufug.2015.06.009. Young, R.F., 2011. Planting the living city: best practices in planning green infrastructure – results from major U.S. cities. J. Am. Plann. Assoc. 77, 368–381, http://dx.doi.org/10.1080/01944363.2011.616996. Zhang, C., Qiu, F., 2012. Mapping individual tree species in an urban forest using airborne lidar data and hyperspectral imagery. Photogramm. Eng. Remote Sens. 78, 1079–1087, http://dx.doi.org/10.14358/PERS.78.10.1079.