Using Size Class Distributions of Species to Deduce the Dynamics of the Private Urban Forest

Private and Public Tree Comparison

The private garden tree estate constituted a much higher percentage of individuals (78%) than in the few other cities for which data are available (Clark et al. 1997; Heynen and Perkins 2005). The high proportion of large trees on private land in the current sample emphasizes the importance of understanding the dynamics of the private tree estate.

Size Class Analysis as a tool to Deduce Urban Tree Dynamics

Urban trees in private gardens are less limited by soils and climate than in natural systems, as gardeners can and do provide supplementary water and nutrients as well as, in some cases, sheltering of young trees from common causes of mortality, such as severe frost and drying winds. This nurturing could be expected to result in lower mortality in age classes, resulting in much flatter reverse-J curves than in a natural forest, as could the lesser competition for space.

Given their nurturing, if garden trees are left to grow as they will, height is a good surrogate for age. However, height is subject to modification by the gardener, who may remove the tops of trees for various reasons, and many of whom use tree taxa for hedging and topiary, although both phenomena are uncommon. Taxa pruned for these purposes include Cupressus macrocarpa, Photinia glabra, and Camellia species. In the cases of the self-seeding taxa, such as Cotoneaster spp., Pittosporum undulatum, and Prunus spp., their voluntary nature and their tendency to be a nuisance in adjacent bushland can lead to an early demise (Zagorski et al. 2004). This truncation of their potential lifespan may concentrate their incidence in the classes of lesser stature. The recent tendency to plant advanced tree stock also complicates interpretation of the data set.

The fact that there was no relationship between growth rate and height class distribution removes one possible difficulty in the interpretation of the results. Small proportions of individuals in the smaller height classes can therefore be most reasonably interpreted as a recent failure to plant potentially tall trees rather than the result of any particularly rapid transition from seedling to 8 m tall tree.

The Changes

The size class distribution data strongly suggested that the private urban tree estate will change in its species composition over the coming decades. The study authors are confident that these changes are real because there are many species of trees that lack, or almost lack, representation in the smallest height class, and many others that lack, or almost lack, representation in the largest height class. It is inferred that the former species are not being planted at a sufficient rate to maintain their present density as mature trees, and that the latter will become more prominent in the tallest class in the future.

A prime example of species in the former class is the slow-growing Liquidambar styraciflua (Figure 2). The general pattern is a downward trend for the potentially tall exotic trees, such as the deciduous trees, Larix spp., Quercus spp., Ulmus spp., Fraxinus spp., and Liquidambar styraciflua, and the evergreen Picea spp., and Pinus spp. (Figure 2). The native rainforest tree Grevillea robusta is also not prominent in the smaller height classes. Tall forest trees have given way to glossy-leaved small trees, such as Magnolia grandiflora and Michelia species. Although some of these species are widely used as hedging plants, most are also planted as specimen trees, and some, such as Michelia, were not observed by us in hedges.

None of the four groups of trees possessed a reverse-J curve, on average. Although trees in group 4 came closest, their greatest density is in the second smallest class rather than the smallest (Figure 1). The patterns in all four groups are typical of the patterns shown in episodic recruitment in natural forests (Figure 1).

Explaining Private Tree Preferences

Researchers postulate the following possible explanatory factors for the observed shifts in private tree species composition: the vagaries of fashion consequent on the increased influence of garden-related industries and popular media on private tree choices; shifts in cultural meanings invested in urban trees ranging from issues of belonging to perceptions of risk; biophysical drivers, such as drought and diseases; and changes to housing trends and urban form, leading to reduced provision of private garden space.

In the 1950s, the ornamental trees for front yard display that were most favored by homeowners and stocked by nurseries in Melbourne were Liquidambar sp. and Betula sp., while the Cupressaceae family and Photinia robusta were used for tall hedges (Kirkpatrick 2006). Today glossy-leaved evergreen trees, preferably with spectacular or interesting flowers or fruits (e.g., Acmena spp. Magnolia grandiflora, Michelia doltsopa) or plants with unusual foliage color (e.g., Pittosporum tenuifolium) are being used as accents. This shift is likely to have been influenced by the increasing scale of garden-related economic consumption over the past fifty years allied to the growth of private garden industries, chiefly nursery, maintenance, landscaping, and arboricultural businesses.

In the second half of 2009, the Australian garden nursery and garden maintenance industries alone accounted for a turnover of AUD 3.8 billion (NGIA/HA 2010). The increased economic importance of garden consumption has been accompanied by increased investment in tree-related advertising and significant growth in gardening content in popular media, such as television programs and magazines (NGIA/HA 2010). While responding to resident tree species preferences, this market activity explicitly seeks to influence consumer demand and, in particular, to promote novelty and changing fashion as mechanisms of sustaining the consumption of tree ‘products’ in the face of the longevity of many trees. This activity raises the prospect that many healthy trees may be removed at a frequency dictated by the vagaries of fashion (Kirkpatrick et al. 2012). It is interesting to note that currently the most highly fashionable trees, those in group 4, are among those that demand the greatest amounts of water, suggesting that such fashion is not necessarily attuned to environmental conditions. This finding is indeed remarkable in the Melbourne context, as the city experienced prolonged drought through most of the first decade of this century, leading to water restrictions, increased water prices and government education programs aimed at promoting ‘water-wise gardening.’ Interestingly, the desire for water efficiency in plants is conflicted by a converse preference for broad leaved, dense canopy trees, which, due to their shade and high levels of evapotranspiration, cool cities (City of Melbourne 2011).

In cities and elsewhere, trees are carriers of cultural meaning (Jones and Cloke 2002). Thus, changes in tree preferences are likely to reflect underlying cultural shifts. The finding that recent tree plantings (group 4) are likely to provide better privacy than earlier, tall-tree plantings is in keeping with the decreasing cultural importance of gardens as bearers of cultural conformity (see Timms 2006) and their increasing importance as sources of privacy and as sites in which to express individuality (Daniels and Kirkpatrick 2006a; Kirkpatrick et al. 2009).

The intensifying of a risk avoidance and public liability culture over the past twenty years in Australia has reduced the attractiveness of large (especially brittle) trees to residents, not least because of increased maintenance costs. Planting guides suggest that cypresses and eucalypts unpredictably drop large limbs, while paperbarks and wattles are also prone to collapse at a relatively young age. Additionally, the impact of the Victorian bushfires in the summer of 2009 and the subsequent amendments to the planning scheme as a result may have influenced homeowner perception of fire risk from trees (Kirkpatrick et al. 2012).

Perhaps the most significant ways in which trees bear cultural meaning in Australia is in relation to questions of place and belonging (Holmes et al. 2008), questions that run deep in colonial societies (Kirkpatrick 2006). The latter part of the twentieth century saw a surge of expression of nativism in urban gardens in Australia (Timms 2006; Head and Muir 2007). Some gardeners restricted their tree palette to natives, or local indigenous species, while others used natives in combination with exotic trees (Daniels and Kirkpatrick 2006a; Kirkpatrick et al. 2007; Kendal et al. 2012). This shift is reflected in the data, particularly in the prevalence of bottlebrushes (Callistemon spp.), native trees that flower profusely, do not grow too tall, and rapidly form dense, obscuring foliage (see Appendix). Habitat gardening is an emerging private garden ideology directed toward the succor of native birds, reptiles, frogs, and mammals (Grant 2003). Large trees provide better and more habitat for many native species than an equivalent area of smaller trees, shrubs, or lawn (Stagoll et al. 2012). Although native animal species use many exotic trees for food and shelter (Low 2003; Daniels and Kirkpatrick 2006b), habitat gardeners tend also to be nativists of the local variety. However, habitat gardeners are too rare in suburbia to have much influence, judging by the very small incidence of complex native gardens in Hobart suburbs (Kirkpatrick et al. 2007).

Changing environmental values have also seen increased uptake of chemical-free, productive gardening in Australian cities, sometimes under the banners of organic gardening or permaculture (Timms 2006). This change has recently seen the longstanding habit of planting fruit trees in Australian backyards extended to front gardens (Kirkpatrick et al. 2007).

Several biophysical factors are likely to be implicated in the observed tree species composition. The droughts and heat waves of the 1990s and 2000s resulted in morbidity and mortality of a wide variety of trees (Kendal 2011). For example, personal observations on the part of the authors suggest that there was widespread mortality of Camellia spp. and Betula spp. in Melbourne during the February 2009 heat wave, and the city’s dominant CBD tree, Platanus, suffered greatly (City of Melbourne 2011). Pests and pathogens have also led to the loss and reduced planting of some species (e.g., Ulmus spp. in Melbourne, Acacia iteaphylla in Hobart). The recently introduced myrtle rust may threaten a large proportion of the garden trees of Melbourne, particularly natives. Pine and cypress trees may have become unpopular because they provide excellent breeding habitat for the native brush-tailed and ring-tailed possums, which in turn feed on suburban fruit trees, exotic ornamentals (e.g., roses and camellias), and vegetable crops (Kirkpatrick 2006). These biophysical factors highlight a vulnerability that comes with relying on a few species of trees.

The trend toward smaller urban trees is also a reflection of the changing physical structure of the urban environment. There has been a marked increase in the residential density of Australian cities over recent decades (Beeton et al. 2006). This closer settlement has been achieved with the subdivision of existing suburban plots and the building of new houses on smaller lots. Coupled with a significant increase in the size of Australian homes, these changes have seen the extent of private garden space reduced considerably (Hall 2010). Combined with heightened interest in risk avoidance, this trend toward smaller gardens militates against the planting of tall trees. The recent uptake of rooftop solar power systems in suburbs and the growing trend for vegetable gardening are likely to further contribute to the planting of smaller trees.

The trend toward shorter stays in individual residences, whether renting or owning, is also likely to contribute to a shift away from slower growing, larger, and long-lived tree species. Given the eight-year mean turnover of house residency (ABS 2009), planting a tree that takes decades to reach a substantial height requires selflessness and foresight.

Planning and Management Implications

Changes in the species composition of the private tree estate reported here have important implications for management efforts to optimize the social and environmental benefits afforded by the urban forest. The finding that large trees are likely to be less common in private gardens than in the past is especially significant as large trees are widely regarded to deliver a disproportionate share of urban forest benefits (Moore 2009). For example, large trees tend to become independent of watering by humans as their roots gain access to groundwater, provide vital urban habitat for many native animal species, and may figure more prominently than small trees in social processes of place-making (Wolf 2005), as they are highly visible in the landscape and are experienced as shared characteristics of a local place. Moreover, the challenging conditions and space constraints experienced by street trees mean that unless new and novel opportunities to plant large trees in public spaces (such as in the roadway) are afforded, the population of large trees will continue to decline. The canopy cover targets that are commonly used to indicate urban forest benefits by municipal tree managers will be considerably harder to deliver with a smaller private tree stock.

An important management context for such canopy cover targets is the potential for anthropogenic climate change to exacerbate urban heat island effects (e.g., City of Melbourne 2011), thereby posing risks to public health and well-being, built infrastructure, and urban biodiversity, amongst others. In this context it needs to be noted that the large deciduous trees in group 3, if properly situated in relation to adjacent housing, are known to moderate extremes of temperature and humidity and thereby lower domestic energy consumption. The replacement of these trees by the smaller species in group 4, along with the potential heating effects of increases in the built fabric, raises the prospect of positive feedback between increased domestic energy consumption, global climate change, and local urban heat island effects (Hamin and Gurran 2009; Kendal 2011). Additionally, carbon storage by large trees is up to 1000 times greater than that of small trees (McPherson et al. 1994). The shade provided by large canopy trees, both on public and private properties where the tree overhangs public space, modifies the local microclimate, reducing temperature and encouraging pedestrians and cyclists, thus influencing public health (City of Melbourne 2011).

Since the early 1980s, municipal governments in Hobart, unlike the majority in Melbourne, have employed UK-style tree preservation orders as a way of protecting larger trees on private as well as public land. The data does not directly answer the question of whether and to what extent healthy large trees have been removed from private gardens; the data does, however, suggest that such regulations may not have been effectively enforced in Hobart. It is possible that public resentment toward local government controls over private trees may be a factor in encouraging residents to plant smaller trees or to remove trees to pre-empt restrictions. In this context, researchers note that the City of Melbourne is currently developing a register of exceptional trees on private property with the intention of protection through the Melbourne Planning Scheme. The study authors also note that regulation has been shown to be effective in retaining large private trees in at least one western city (Sung 2012).

While managers have a growing range of regulatory instruments at their disposal, the mandated planting of certain types of trees on private land is politically difficult to advocate and virtually impossible to implement. In this context, one management implication of the data is that loss of tall trees in the private estate needs to be compensated for by planting on public land. For example, if the Norfolk Island pines and Canary Island palms that characterize coastal suburbs in Melbourne begin to disappear from gardens, they can be preferentially planted in streets and reserves by municipal land managers. The difficulty here is that use of public space is increasingly contested in Australian cities. Urban consolidation through subdivision of existing blocks, for example, typically reduces the viable room for street trees on the adjacent street verge through the requirement for multiple driveways. Urban consolidation in the form of new developments also typically has lower public space provision as well as smaller setbacks of housing from the street than was the case in most of the twentieth century. Finally, while some Australian suburbs have increased overhead space for large trees by burying power lines, competition for underground space has intensified, not only due to the addition of power lines, but as a result of increased residential densities and the presence of new telecommunications services.

Regulatory approaches are only one way in which public managers seek to influence the tree management practices of private residents. One alternative is to support tree-related industries and councils in educating the public about the systemic benefits of the urban forest and to ensure that there is adequate availability of appropriate tree stock. Related to this is the opportunity to raise public awareness of the invaluable contribution of private trees to the urban forest through education and the celebration of existing trees. This could be supported by increased popular media content about gardening. Recognition of the values of existing trees through such mechanisms as council registers, urban forest art and design competitions, or walking tours may assist in this goal. Furthermore, by offering incentives for the retention of existing trees, such as free arboricultural advice, the contribution of private tree estate to urban forest function may be maintained. Another alternative is to directly engage residents in urban forest management on public land. The potential for such involvement may be particularly high in Australia, where horticultural societies and gardening groups have a strong presence. Community Landcare groups have partnered with government to rehabilitate public bushland over the past few decades (Davison and Ridder 2006). There is little research examining the relationship between participation in these groups and private garden practices. The potential of these groups to raise the awareness of residents about the importance of the private tree estate in urban forest management is not known.

One difficulty facing managers in seeking to educate residents about the importance of their role as private land managers within the urban forest is the dearth of research into the motives, and especially the actions, of private residents in relation to urban trees. Such research may not only identify effective educational and regulatory strategies, but also reveal ways in which private residents are uniquely positioned to contribute to the well-being of the urban forest. Consider, for example, the emerging emphasis on increasing tree species diversity in Australian urban forest management, as a strategy for building resilience in the urban forest in the face of pests, diseases, and environmental change more generally (City of Melbourne 2011; North Sydney Council 2011). In this regard, an observed increased cultural emphasis on individual self-expression through gardening (Kirkpatrick et al. 2009) may be a mechanism by which the species diversity of the urban forest may be maintained. The high proportion of tree cover found in the private realm, and the changing nature of that tree cover as exposed by this study suggests a need to understand the current composition and influencing drivers of private trees. By carrying out collective management of the urban forest that considers the private tree estate, the social, ecological, and environmental benefits afforded by the urban forest will be enhanced.