Drivers of diversity and tree cover in gardens, parks and streetscapes in an Australian city
Introduction
It is widely recognised that urban vegetation is not distributed equally (Talarchek, 1990, Iverson and Cook, 2000, Heynen et al., 2006). This leads to inequality in the provision of important benefits that urban vegetation provides to people such as ecosystem services (Bolund and Hunhammar, 1999), improved health and wellbeing (Mitchell and Popham, 2008) and reduced crime rates (Kuo and Sullivan, 2001), and to the urban environment such as the provision of food and habitat for other organisms (Angold et al., 2006). The distribution of urban vegetation is influenced by government policies such as those directing planting in public open space and streetscapes, and by those regulating the removal of trees from private property (Conway and Urbani, 2007). Greater understanding of the drivers of the distribution of urban vegetation in different cities will aid development and implementation of suitable policies to address inequality in the distribution of urban vegetation (Kirkpatrick et al., 2011), and consequently reduce inequality in the provision of by the urban forest (Landry and Chakraborty, 2009).
While there are a number of studies that show that socioeconomic factors drive inequality in the distribution of urban vegetation, the diversity of places studied and the range of distribution measures used has been limited. Empirical studies have been mostly limited to a single measure of distribution (typically remotely sensed tree cover), generally in North American cities with high levels of socioeconomic inequality, and mostly of a single land use (or an aggregation of multiple land uses). Comparative studies in places with different social and physical environments, using alternative measures of distribution are required to test whether generalisations can be made from the theories that have been proposed to explain observed patterns (Talarchek, 1990, Kinzig et al., 2005).
A number of competing theories have been proposed to explain the observed inequality in the distribution of urban vegetation. The first group of theories relate to the importance of different socioeconomic drivers of distribution. Household income has been identified as an important driver of both tree cover in New Orleans, USA (Talarchek, 1990) and Chicago, USA (Iverson and Cook, 2000) and vegetation diversity in Phoenix, USA (Hope et al., 2003) and Tlokwe City Municipal area, South Africa (Lubbe et al., 2010). The importance of income in determining the distribution of vegetation has been characterised as a ‘luxury effect’ based on social stratification, where people with the economic wherewithal move to landscapes with higher plant diversity (Hope et al., 2003), or alter their local environment by planting trees (Mennis, 2006). At the same time, low income households face barriers to the presence of vegetation such as the lack of available space or limitations on property control that may forbid new plantings (Talarchek, 1990), and may preferentially cultivate a limited range of economically useful species (Lubbe et al., 2010). In contrast, education level is a better predictor of tree cover than income in central Indiana, USA (Heynen and Lindsey, 2003), Baltimore, USA (Troy et al., 2007) and regional cities in south-east Australia (Luck et al., 2009). This has been explained by an increased knowledge of the benefits of living with trees leading to an increased likelihood of choosing to live in areas with existing trees or to undertake activities such as planting trees (Luck et al., 2009). Research in environmental psychology provides some support for this, showing that people with higher education levels value trees more highly (Lohr et al., 2004). A third approach has argued that different social groups have different landscaping tastes (Whitney and Adams, 1980, Talarchek, 1990), and that group identity associated with lifestyle behaviour (determined by factors such as life stage, family size and ethnicity in addition to socioeconomic factors) is a better predictor of tree cover than socioeconomic factors alone (Grove et al., 2006).
The second group of theories relate to the type of landscape where this inequality is observed. Urban ecology studies have argued that the ‘luxury effect’ leads to ‘bottom up’ effects, where the integrated outcome of many individual actions and choices at the household level reflect social differences between decision makers in different neighbourhoods (Martin et al., 2004). These ‘bottom up’ effects lead to greater inequality in private landscapes such as residential gardens. This is contrasted with centralised ‘top down’ planning and management influences which result in vegetation being distributed more evenly in public landscapes. In contrast, political ecology and environmental justice research has argued that the distribution of public trees are influenced by political processes that tend to reflect socioeconomic inequality, and that inequality will be more prevalent in public landscapes such as streetscapes (Heynen et al., 2006, Landry and Chakraborty, 2009). For example “it is still expected that more effort is taken to publicly subsidise and maintain street trees within upper-class and middle-class areas owing to the political responsiveness and social-capital accumulation that tends to correlate to middle-class income areas” (Heynen, 2006, p. 512). These contrasting positions are both supported by empirical evidence, albeit in different places and using different measures of distribution (diversity and tree cover respectively). This study explores this apparent contradiction by exploring both diversity and tree cover in public and private landscapes within a single municipality. Greater understanding of the drivers of inequality can help direct actions and public policy designed to reduce it.
There are several limitations with the existing research on the distribution of urban vegetation. Most studies have been undertaken in cities with high levels of income inequality in the Americas and South Africa. In these cities, income and education level are often highly correlated (e.g. Landry and Chakraborty, 2009) making it difficult to untangle the influence of income and education. As a consequence, many studies have included income but not education level in their analysis (Talarchek, 1990, Iverson and Cook, 2000, Jensen et al., 2004, Conway and Hackworth, 2007) and findings have usually been related to income alone. Several important studies have been undertaken in the semi-arid desert city of Phoenix (Hope et al., 2003, Martin et al., 2004) which has a climate that is hotter and drier than many other western cities. Climate has been shown to be an important driver of urban vegetation (Kendal et al., in press) as urban tree cover tends to be related to tree cover in the surrounding landscape (Dwyer et al., 2000). Comparative studies are required to see whether patterns observed in Phoenix are also observed in temperate, tropical and cold climate cities. Another important study by Grove et al. (2006) used a commercial marketing database with indicators that do not have readily available equivalents outside the USA to show that models including ‘lifestyle behaviours’ are better predictors of tree cover than models including socioeconomic factors alone. However, the significance and relative importance of individual variables, and the size and direction of effects were not reported, which prevents comparison using disaggregated census data available in other places. Lastly, by far the majority of studies have been conducted using remotely sensed tree cover, a relatively easily and widely available dataset. However, there are other important measures of vegetation distribution such as species richness that are not necessarily related to tree cover. Species richness is an important property of ecological systems which can contribute to the resilience of the biological component of the system (Folke et al., 1996), influences people's health and wellbeing (Fuller et al., 2007) and help shape people's experience of biological life (Miller, 2005).
This study explores the relative importance of household income and education level on patterns of vegetation distribution across the urban area of Ballarat, Australia using two different measures of distribution (tree cover and species richness). These patterns are also explored across different land uses (private gardens and public parks and streetscapes), which allows us to explore how relevant different theories on the drivers of vegetation inequality are in the context of an Australian city; it is anticipated that the relative importance of individual behaviours (‘bottom up’ process) will be revealed in private landscapes and political (or ‘top down’) processes in public landscapes. The city of Ballarat is a useful place to study drivers of vegetation distribution as there is variation in both tree cover and species richness, and physical and social variables, and it is governed by a single municipal authority which excludes the potentially confounding influence of different regulatory and management regimes that may occur across municipal boundaries. Results will be compared with other studies to explore broader patterns in the relative importance of these variables on the distribution of urban vegetation.
Section snippets
Study site
Ballarat is a regional city that is centrally located (latitude 37.56 S, longitude 143.86 E) in the state of Victoria. It is a relatively old Australian city founded during the gold rush of the 1850s. It is one of the largest inland cities in Australia with a population of approximately 82,000 people and has a high degree of socioeconomic inequality, with census areas ranging from the 1st to the 89th percentile of the Australia-wide Index of Relative Advantage and Disadvantage (IRSAD) (
Drivers of tree cover on residential property, road reserves and parks
The residential property in 156 census districts, residential road reserves in 154 census districts (two census districts contained residential property but no road reserves) and parks in the 86 census districts were analysed (see Appendix A for details). Areas with a high proportion of residents with a graduate education had more tree cover across both public and private land uses (Table 3 and Fig. 3). Household income was not included in any best model. Areas with older housing also had
Tree cover
It is clear that in Ballarat, some people with higher incomes and relatively low education levels are living in new housing estates with low levels of tree cover on both private and public land, rather than areas with established housing and higher levels of tree cover. This does not support the idea that people with high incomes preferentially choose to live in treed neighbourhoods (Hope et al., 2003, Martin et al., 2004). It may be that other factors found in newer neighbourhoods such as the
Acknowledgments
The authors would like to thank City of Ballarat for access to aerial photography, the residents of Ballarat who consented to the vegetation survey of their front yards, and several anonymous reviewers whose comments have contributed to a greatly improved manuscript. This research was funded by an Australian Postgraduate Award.
References (62)
- et al.
Biodiversity in urban habitat patches
Science of the Total Environment
(2006) On the measurement of inequality
Journal of Economic Theory
(1970)- et al.
Ecosystem services in urban areas
Ecological Economics
(1999) - et al.
Variations in municipal urban forestry policies: a case study of Toronto, Canada
Urban Forestry & Urban Greening
(2007) - et al.
Landscapes of the lifespan: exploring accounts of own gardens and gardening
Journal of Environmental Psychology
(2007) - et al.
Temporal and spatial variation in garden and street trees in six eastern Australian cities
Landscape and Urban Planning
(2011) - et al.
Integration of landsat thematic mapper and census data for quality of life assessment
Remote Sensing of Environment
(1997) - et al.
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) Biodiversity conservation and the extinction of experience
Trends in Ecology & Evolution
(2005)- et al.
Effect of exposure to natural environment on health inequalities: an observational population study
The Lancet
(2008)