Adaptations of Australian Tree Species Relevant to Water Scarcity in the Urban Forest

THE PRIORITY FOR WATER AND THE URBAN FOREST

While urban forests are beautiful and decorative, these attributes often conceal the many functions and services that they provide to cities to the point where their social, health, economic, and environmental benefits are overlooked (McPherson 2007; Moore 2009; Nowak et al. 2010). What else delivers so many benefits immediately, and benefits that last centuries into the future, prolonging healthy lives and making cities both sustainable and liveable? Urban forests have been silent assets to cities for decades and even centuries. They are major and essential urban infrastructure (Daniels and Tait 2005).

Cities are biodiversity hot spots due to the variety of habitats available in public and private open space, especially the diversity of plantings in domestic front and back yards (Daniels and Tait 2005). The requirement for tree managers is to establish a priority for the urban forest in the allocation of a precious and valuable, rather than scarce, commodity (Connellan 2008). Society will allocate water to items for which there is an economic and political imperative.

For most of its history, the price of water in Australia has been subsidized, however, it does have a real economic value and in most States increasing water prices are moving toward that value (Victorian Department of Sustainability and Environment 2004; Victorian Department of Sustainability and Environment 2006; Victorian Department of Sustainability and Environment 2007). To maintain the urban forest, water must be used effectively and efficiently. There can be no going back to the days of profligate water use and year-round emerald green lawns (Moore 2009). The environment and economy cannot sustain such an approach (Water Resources Strategy Committee 2002). How well informed are the practices governing the use of water in the urban forest and what are the research needs that would enhance best management practices?

ADAPTATIONS RELEVANT TO WATER STRESS

Trees in the urban forest face the dilemma of all terrestrial plants: the need to balance the interaction of carbon and water cycles to allow survival and growth. If water is limited and stomata close, carbon assimilation through photosynthesis is reduced (Cowan 1981; Curran et al. 2009; Martin St. Paul et al. 2012). Thus in the urban environment, restricting water availability to trees in the urban forest may also restrict the benefits that they provide, such as their capacity for carbon sequestration (Jonson and Freudenberger 2011) and transpirational cooling.

The performance of different trees species in minimizing water loss, but at the same time maintaining carbon dioxide gain, is defined as water-use efficiency:

The value of water use efficiency varies for different species and can be used to select trees that are more productive for use in cities of drier climates (Ladiges et al. 2005).

Australian tree species possess many and varied adaptations to growing in arid environments (Table 1). One of the defining characteristics of many Australian plant genera is sclerophylly. Sclerophyllous trees possess large amounts of sclerenchyma tissue, which maintains cellular volume as conditions dry. It is often assumed that sclerophylls are low water users, but paradoxically many have poor stomatal control and will use whatever water is available until they wilt (Ladiges et al. 2005). Many have the capacity to survive in environments where water is limited, and managers could proactively minimize the supply of water in low-water environments using sclerophyllous trees.

The leaves and phyllodes of many Australian species (Table 1) are isobilateral and often hang vertically, thereby reducing the surface area that is exposed to the sun (King 1997). Species such as Eucalyptus preissiana (Knox et al. 1994) and E. obliqua have prominent cuticular ledges, which overarch their stomata, creating a stomatal antechamber that reduces transpirational water loss (Moore 1981). However, the stomatal anatomy of many common street trees species remains unknown.

In Australian tree species, the number of stomata ranges from about 28 mm⁻² in Persoonia (geebung) to between 100–350 per mm² in eucalypts. The number often varies inversely with size with fewer larger stomata contrasting many smaller stomata (Knox et al. 1994). In Eucalyptus globulus, there are 300 stomata mm⁻², but the leaf area occupied by stomatal apertures is only about 1%. However, with stomata open, the rate of transpirational water loss is the same as for evaporation from an open wet surface; water and gaseous movement through open stomata is remarkably efficient. Thus, knowledge of stomatal rhythms and behavior is essential to understanding tree water use and survival in water-limited environments.

Trees such as Casuarina littoralis, Eucalyptus calophylla, Eremophila macgillivrayi, Pittosporum phylliraeoides, and Myoporum floribundum show effective stomatal control and so more efficient water use, but if water is limited then their growth rates may be slowed to the point where they are ineffective for planting in the urban forest. Similarly, species such as Acacia melanoxylon or Eucalyptus grandiflora, which reduce water use through reduction in leaf surface area, may lack the canopy characteristics and density that would make them attractive for urban forest planting.

For most Australian tree species planted in urban environments there are almost no data on basic physiological processes, such as stomatal behavior, let alone whether they are stress avoiders or tolerators in relation to water (Table 2). Which trees have good stomatal control as soil moisture diminishes (Eamus et al. 2001; Prior et al. 2005), which keep their stomata open and so are luxury water-users, and which species can tolerate low internal water potentials are largely unknown (Atwell et al. 1999), except for those few species that are of interest for forestry, timber, or agricultural research (Pate and McComb 1981; Meier and Leuschner 2008). Such basic research would not take large amounts of funding, and simple data gathering using basic porometry would not take long, but this has not attracted the interest of the research funding bodies.

Acacia is Australia’s largest indigenous genus with over 900 woody species ranging from shrubs to large trees. They are generally sclerophyllous and Australian species are typically phyllodenous in contrast to the Acacia species of Africa and South America (Thukten 2006). Many arid zone Acacia species are known for their extreme avoidance of desiccation (New 1984; Broadhurst and Young 2006; Page et al. 2011). While A. harpophylla is more drought resistant than A. aneura, even the latter has phyllodes that can lose a large proportion of their water content without harm.

Many species maintain cell turgor despite high levels of moisture stress. In some species, phyllode size reduces in drier areas (Thukten 2006; Deines et al. 2011). The size and shape of A. melanoxylon phyllodes are affected by both aridity and seasonal rainfall patterns (Farrell and Ashton 1978). Several Acacia species have very deep roots that may reach depths of 12 m or more (Table 2). A. mearnsii may have roots that penetrate to 6 m, but 75% of the root system is within 600 mm of the soil surface.

The closure of pinnules as soils dry is easily observed in A. mearnsii—a bi-pinnate leafed species—growing in the basaltic clays of the western plains near Melbourne. This reduces transpirational water loss. In plantations, A. mearnsii could lose 261 kg of water per day compared to A. decurrens’ 44 kg, but this was largely due to a difference in foliage density with A. mearnsii having a foliage mass of 69 kg, while A. decurrens had a foliage mass of 9 kg (New 1984). In an urban forest, a choice between these species may come down to a decision about canopy appearance, density and impact versus water use.

There are major research gaps in the use of Australian native species, as well as exotic species, growing under Australian environmental conditions. Few studies are available on water use by urban trees growing within the urban environment (Misra and Sands 1993), despite an urgent need by tree and water resource managers for quantification (Connellan 2008). There are better data on the irrigation required for establishing young trees (May 2004).

Drought avoiders such as E. camaldulensis, E. regnans, and E. marginata are profligate luxury water-users that will grow rapidly and use significant volumes of water if it is available. They may be inappropriate for urban use where water is limited in supply or costly, while proving ideal for places where water is abundant or as part of water-sensitive urban design measures to control local flooding by holding and absorbing water during more intense rainfall events predicted under a changed climate (Killicoat et al. 2002; Moore 2009). The economic value of reducing localized flooding could be substantial (Moore 2009). Research shows trees to be effective in removing pollutants, such as nitrogen and phosphorus, from stormwater run-off (Denman 2006), and may prove to be useful, long-term elements of water-sensitive urban design.

Many tree species also possess physiological, anatomical, and morphological adaptations to growing in arid conditions (Kursar et al. 2009). Many eucalypt species seem to remain physiologically active, using water under conditions of moderate to severe water stress, reflecting their mesophytic evolutionary origins. However, not all eucalypts are equal in their capacity to cope with dry conditions. In Western Australia, E. calophylla has better stomatal control than E. marginata, which is a luxury water-user. Similarly, in eastern Australia, E. regnans is a profligate water-user with little capacity for stomatal control, while E. obliqua behaves similarly to E. calophylla.

It is interesting to compare a hypothetical scenario where Pinus radiata and Eucalyptus rossii are planted in the same, low phosphorus Australian soil in an urban streetscape where rainfall is low and there is no irrigation after the first year of establishment. When soil water potential falls, the P. radiata closes stomata, reducing photosynthetic assimilation and growth. The E. rossii on the other hand keeps stomata open and tolerates a decline in internal water potential. When occasional light rain falls, the E. rossii resumes photosynthetic assimilation immediately and commences growth (Florence 1981). The P. radiata does not open its stomata and the soil dries, perhaps compounded by the opportunistic uptake of water by E. rossii. The E. rossii out grows and out competes the P. radiata under this scenario.

Winter deciduous Australian native trees are relatively rare, with Melia azedarach, Nothofagus gunnii, and Brachychiton acerifolius being notable examples. Furthermore a few northern species, including some eucalypts, such as E. clavigera, E. grandiflora, and E. brachyandra, are facultatively deciduous during the dry period (Williams et al. 1997). This characteristic is shared with a number of other tree species, some of which are suitable for urban use (Table 3). However, there has been very little breeding and selection of these native species for urban use, and even less research on whether breeding might allow deciduousness to apply to southern winters, expanding the potential use of any of these or related species (Munne-Bosch and Alegre 2004).

Some species have stomata that respond to the vapor pressure of the ambient air (Table 1). Stomata close in response to drier air and leaf moisture content increases as a result, but transpiration reduces accordingly. Species with this characteristic could prove very useful in cities where water is limited, but while the response has been observed in some species with potential for urban use, it is largely unresearched.

Some species of Australian urban trees come from populations that have wide and extensive natural distributions in environments where water availability varies (Wheeler et al. 2003). There are good data to inform provenance selections for many forest species (Hamrick 2004; Broadmeadow et al. 2005; Craft and Ashley 2007; Gouveia and Freitas 2009), but arboricultural data on Australian species of amenity trees are not so easily accessed. Studies on provenances of Lophostemon confertus (Williams 1996) and Tristaniopsis laurina (Looker 2001), from different climate and soil conditions, have been undertaken and would allow urban selections for drier climates. Even if species’ ranges are limited, there may be the option of selecting different species from within a genus. This is the case with the genera Eucalyptus and Acacia within Australia, where there are large numbers of related species occupying a broad range of habitats.

Often in eucalypt-dominated forests it is common for different species to occupy environments that become increasingly drier (Fensham and Holman 1999). This gives rise to the concept of a displacement series, of often-related species, which replace each other over an ecotone of increasingly arid environments (Pate and McComb 1981; Shepherd et al. 2008; Holman et al. 2011). As this happens, species have a tendency to show characteristics (Table 4) that better adapt them to the drier conditions. These characteristics could be used by urban forest managers as a guide for what species might be successful for urban planting in drier conditions, but very little research has been applied to the urban context.

Good Australian data support the use of irrigation under singular mulches in general, and mixed particle size organic mulches in particular (Connellan et al. 2000; Handreck and Black 2002). Early morning subsurface irrigation regimes that permit trees to open stomata early to maximize photosynthesis before water becomes limiting are based on sound tree physiology. In many species, stomata are often closed by about 2:00 pm, especially if soil water is limiting (Eamus 2006). Furthermore, for many tree species evapotranspiration cools them, reducing the risks of heat damage, especially on hot windy days, the frequency of which is likely to increase under climate change. Such irrigation also captures at least some of the general and environmental benefits that the urban forest provides in terms of transpirational cooling.

ROOT ARCHITECTURE AND WATER USE

When a tree seed germinates in natural soils, the radicle emerges and usually develops into a tap root. In Australian native tree species, such as Eucalyptus and Acacia, it is not uncommon to find a seedling of 20 mm height with a primary root of 150–200 mm in length (Moore 2008). This root then rapidly develops as a tap root, anchoring the young tree, providing necessary water and nutrients and the framework from which lateral roots develop (Awe et al. 1976). In most urban trees, however, the tap root should be considered a juvenile characteristic, which only persists for the early establishment phase of the tree’s life cycle (Ashton 1975; Moore 1990).

The root systems of mature trees have a tendency to be spreading and relatively shallow (Watson and Neely 1994). The typical urban forest tree root system consists of a shallow spreading root plate of lateral spreading roots complemented by the presence of descending (or vertical or sinker) roots, which usually occur around the base of the tree or close to the trunk, where oxygen is more readily available and where nutrients and organic matter are being actively recycled (Coile 1937; Perry 1982). While the lateral roots are often within 200–300 mm of the soil surface, descending roots may grow to depths of 1000 mm or more. There are also descending roots farther out along the root plate, which have a tendency to be smaller in diameter and shallower in their descent. These roots may persist for a number of years before they die back and are replaced (Moore 1995; Smith and Moore 1997).

This common pattern of urban tree root architecture has profound implications for the application of water. However, there are few data on the variations in root architecture for native and exotic trees and almost none comparing Australian native species. Many irrigation regimes assume that roots are close to the trunk and under the drip line of canopies. This seems to be the case for species such as elms, but is not necessarily the case for eucalypts and other species where exposure of root systems with an air knife shows the presence of major structural roots within the drip line but very few, if any, fine absorbing roots (Moore 2008). The absorbing roots are often 10 m or more from the trunk and concentrated where moisture levels are higher.

There is an urgent need for data on the root architecture of Australian urban tree species. It is vital to know where roots are, why they develop where they do, and how much water they are capable of removing from soil in their vicinity. It is also essential to know where, and at what depth, water should be supplied for efficient and effective irrigation (Connellan 2008). There is a popular view that trees absorb water from deep in the soil profile and that only “deep soaking” is effective irrigation over summer. Current knowledge of root architecture suggests that this is not the case for urban forest trees, but there is little research to inform the debate. Consequently, water restrictions that limit irrigation of urban trees have been imposed rather than allowing an occasional irrigation of the absorbing root plate near the soil surface. This has resulted in higher levels of stress and the deaths of many mature trees in the urban forest over the past decade.