Scattered trees: a complementary strategy for facilitating adaptive responses to climate change in modified landscapes?


*Correspondence author. E-mail:


1. Facilitating adaptive responses of organisms in modified landscape will be essential to overcome the negative effects of climate change and its interaction with land use change. Without such action, many organisms will be prevented from achieving the predicted range shifts they need to survive.

2. Scattered trees are a prominent feature of many modified landscapes, and could play an important role in facilitating climate change adaptation. They are keystone structures because of the disproportionally large ecological values and ecosystem services that they provide relative to the area they occupy in these landscapes. The provision of habitat and connectivity will be particularly relevant.

3. Scattered trees are declining in modified landscapes due to elevated tree mortality and poor recruitment often associated with intensive land use. The continuing global decline of scattered trees will undermine the capacity of many organisms to adapt to climate change.

4.Synthesis and applications. The sustainable management of scattered trees in modified landscapes could complement other strategies for facilitating climate change adaptation. They create continuous, though sparse, vegetation cover that permits multi-directional movements of biota across landscapes and ecological networks. They have the capacity to span ecosystems and climatic gradients that cannot be captured in formal reserves alone. The management of scattered trees should be an integral part of conservation objectives and agricultural activities in modified landscapes. Public investment, through mechanisms such as agri-environmental schemes, in rotational grazing, temporary set-asides, tree-planting and regulations that reduce clearing and early mortality among standing trees will improve the capacity of biota to adapt to climate change.


There is a growing body of evidence that rapid climate change is having profound impacts on biodiversity. A major effect is to force species to shift their ranges (e.g. Parmesan & Yohe 2003; Root et al. 2003; Parmesan 2006). Modelling bioclimatic envelopes under different climate change scenarios provides important information on possible range shifts. However, such modelling alone cannot provide guidance about appropriate management responses at specific locations. A major reason for this is the interaction between climate change and land use change. Land use change that is associated with significant habitat modification inhibits movements by many organisms (Peters 1990; Erasmus et al. 2002; Opdam & Wascher 2004). Therefore climate change poses a major challenge for the management of modified landscapes over which organisms will need to move (Opdam & Wascher 2004; Hulme 2005; Harris et al. 2006). There is now considerable literature on the effects of climate change on organisms (Parmesan 2006). However, there is much less information on how to manage landscapes for biodiversity adaptation to climate change.

To date, some suggested approaches to facilitate adaptive responses by organisms to climate change (hereafter termed ‘adaptation’) include: increasing the number of parks and reserves (Halpin 1997; Hannah 2008); assisted colonization and translocations (Hoegh-Guldberg et al. 2008); ex situ conservation (e.g. seed banks, zoos, breeding programmes, Hannah 2008); limiting threats, such as habitat loss (Hannah 2008); and, enhancing connectivity, in particular through ‘ecological networks’ (Opdam & Wascher 2004; Opdam et al. 2006).

It is unlikely that any individual approach alone will be successful in facilitating adaptation, not least because organisms are expected to respond individualistically (Peters 1990; Erasmus et al. 2002; Hannah 2008), and also because most of the Earth’s land surface is outside reserves (Lindenmayer & Franklin 2002; Rodrigues et al. 2004). There is therefore a strong imperative to manage ‘whole landscapes’, both on- and off-reserve, to allow the maximum number of species to respond individualistically to climate change.

Scattered trees are landscape elements that occur in natural landscapes, cultural landscapes and recently modified landscapes worldwide (Manning et al. 2006). They have great potential to facilitate adaptation and to influence ‘landscape fluidity’ (the ebb and flow of different organisms within a landscape through time, Manning et al. 2009). A key defining feature of scattered trees is their dispersed, open pattern in a landscape (Manning et al. 2006). In modified landscapes, scattered trees can often result from the modification of denser or more intact woodlands and forests (Harvey & Haber 1999; McIntyre & Hobbs 1999; Pulido et al. 2001). They can occur as legacies following clearing or thinning of forest or woodland, or they can be maintained incidentally, or deliberately, as part of agroforestry systems (Manning et al. 2006). Examples are ‘wood-pasture’ in the United Kingdom (Peterken 1996) or ‘dehesas’ in Spain and Portugal (Díaz et al. 1997). However, scattered trees are declining in intensively managed landscapes as a result of too little, or too much (i.e. a shift from woodland to dense forest) tree regeneration due to changed management regimes (e.g. fire, grazing), clearing and tree mortality, salinity, poor tree health and agricultural intensification (e.g. root disturbance by machinery, herbicide drift) (Maron 2005; Manning et al. 2006; Gibbons et al. 2008).

The potential for scattered trees to facilitate adaptation – particularly in production landscapes – has yet to be recognized. In a search of the ISI Web of Science for ‘scattered trees’ (or the synonyms: ‘isolated trees’, ‘remnant trees’ and ‘paddock trees’), or ‘wood-pasture’, or ‘dehesas’ combined with ‘adaptation’, we did not find a single article relevant to climate change adaptation for biodiversity (; accessed 13 October 2008). While this quick literature search may have missed some papers in this area (and did not include books and grey literature), the paucity of results illustrates a general lack of thinking about the potential of scattered trees for facilitating adaptation.

In this paper we outline: (i) how scattered trees can be critically important for biodiversity conservation; (ii) how they could act as vital elements in strategies to facilitate adaptation to climate change, including complementing reserves and ecological networks; and (iii) indicate some possible future directions in research, landscape management and policy.

The ecological value of scattered trees

Scattered trees are recognized as ‘keystone structures’, due to the important ecological values they provide (Munzbergova & Ward 2002; Plieninger et al. 2003; Tews et al. 2004), and their disproportionately large effect on ecosystems relative to the area they occupy (Manning et al. 2006). The ecological value of scattered trees includes conservation of soil nutrients, acting as focal points for tree regeneration, increased vegetation structural complexity, and the provision of habitat and connectivity for some biota (Manning et al. 2006). Scattered trees often occur ‘off reserve’, and can have considerable value for biodiversity conservation. For example, in a study of highly modified, formerly forested landscapes in Costa Rica (which included scattered remnant trees), Sekercioglu et al. (2007) found that many forest bird species lived in the agricultural matrix. In another example, some grazed medium density (6–20 trees ha−1) Eucalyptus woodlands in south-eastern Australia have been shown to contain a native plant richness similar to that of denser (>20 trees ha−1), more ‘natural’ areas, without adversely affecting pasture production (Le Brocque et al. 2009). Scattered trees also provide a range of ecosystem services that are beneficial to agricultural production including: shade and shelter for stock (Harvey & Haber 1999), a natural buffer against soil acidity (Wilson 2002), increased local water balance and nutrient concentration (Ludwig et al. 1999), mitigation of erosion and desertification (Plieninger et al. 2004), and habitat for insectivores and pollinators (Lumsden & Bennett 2005). Scattered trees also can be a cost-effective source of seed for revegetation (Dorrough & Moxham 2005).

Traditionally, modified landscapes have been viewed as patches of habitat in a matrix or ‘sea’ of greater or lesser hostility, with varying levels of ‘permeability’ (McIntyre & Hobbs 1999; Haila 2002). Yet habitat is not always destroyed, but rather modified on a gradient or continuum (McIntyre & Hobbs 1999). Also, different organisms perceive, respond and use the same landscape differently so that a hostile ‘matrix’ to one organism can be ‘habitat’ to another (Manning et al. 2004). To some extent, approaches that focus on reserves (and connectivity between them) are underpinned by patch-based landscape concepts (Manning et al. 2004). This patch-based view of landscapes may not be suited to understanding or managing for adaptation to climate change because of individualistic organism responses. Consequently, whole-landscape, gradient/continua concepts may be more appropriate in a dynamic environment.

Scattered trees fit well within gradient/continua concepts of landscapes, and are important landscape elements that allow many organisms to live off-reserve and across whole landscapes. Examples include birds and bats in south-eastern Australia (Law et al. 2000; Fischer & Lindenmayer 2002a,b) and in Central America (Guevara & Laborde 1993; Galindo-González & Sosa 2003). In many scattered tree landscapes, there is a juxtaposition of two habitat types, forest/woodland and grassland. This can create greater levels of biodiversity and the coexistence of different communities in the same landscape (for example, dehesas, M. Díaz, personal communication). The positive effect of ‘habitat heterogeneity’ on biodiversity is well known (Bennett et al. 2006), and may be a particularly valuable property of scattered tree landscapes in future. This is because they may facilitate the adaptation of higher numbers of organisms to climate change.

Scattered trees can make landscapes effectively ‘useable’ for many woodland and forest organisms. Usability can range from stepping stones for movement (i.e. ‘permeability’), providing connectivity for plants (e.g. through seed movement by frugivores; Galindo-González & Sosa 2003) and animals (Guevara & Laborde 1993; Fischer & Lindenmayer 2002b), to provision of foraging habitat, nesting sites or shelter from weather or predators. For many organisms, scattered trees provide all their requirements, and they can complete their life cycle by using them, while others use them some of the time.

Scattered trees and facilitating adaptation to climate change

Many landscapes are an evolutionary novelty to organisms because the environments in which they evolved were considerably less fragmented than modern landscapes (Lima & Zollner 1996). Consequently, when managing landscapes to facilitate adaptation, it would be desirable to mimic the functional properties of less-disturbed ecosystems, such as structural and compositional heterogeneity, self-organization and gradual boundary transitions. This will be best achieved through a whole-landscape approach rather than being restricted to reserves and protected areas.

For landscape managers, some key potential benefits of scattered trees include:

1.Continuous sparse cover. As the interaction of climate change with land use change is a major threat to organisms, there are strong imperatives to create continuous tree cover across landscapes that allow multi-directional movements in response to climate change. Scattered trees could help re-establish key properties of more natural ecosystems (see above) by increasing habitat heterogeneity and replacing abrupt boundaries with more gradual ones and by ‘de-fragmenting’ modified landscapes. As a result, for many organisms, landscapes would become more useable (see above), and scattered trees could facilitate range-shifts, and effective habitat areas for some organisms.

Continuous sparse cover also allows an even distribution of the ecological values of scattered trees (e.g. improved water balance, shade and shelter, foraging space) across whole landscapes. This spread of resources and functions could facilitate range shifts that may not otherwise be possible for some organisms.

2.Complementing reserves and ecological networks. An important large-scale strategy for facilitating adaptation will be a shift from protected area-only approaches towards ‘ecological networks’ including protected areas, corridors and surrounding landscapes that will provide connectivity across landscapes (Opdam & Wascher 2004; Opdam et al. 2006). Opdam et al. (2006) define an ecological network as:

a set of ecosystems of one type, linked into a spatially coherent system through flows of organisms, and interacting with the landscape matrix in which it is embedded. (p. 324).

There are over 150 landscape-scale and regional ecological networks under development around the world (Bennett 2004). For example: Yellowstone to Yukon in the United States (, Great Eastern Ranges Initiative ( and Kosciuszko to Coast ( in Australia, The Pan-European Ecological Network in Europe ( and Mesoamerican Biological Corridor (

While ecological networks will provide the backbone of adaptation strategies, they may not be enough on their own. This is because structural connectivity (e.g. corridors), as perceived by humans, does not always increase functional connectivity for other organisms (Donald & Evans 2006; Lindenmayer & Fischer 2006; Boitani et al. 2007). Thus, the importance of the landscape context surrounding ecological networks extends far beyond management of the ‘permeability’ of the matrix to organisms. For example, Sekercioglu et al. (2007) found forest birds in Costa Rica ‘did not commute from extensive forest; rather they fed and bred in the agricultural countryside’ (p. 482). The condition of the matrix surrounding habitat patches or reserves has a profound influence on ecological communities within fragmented landscapes (see Donald & Evans 2006; Kupfer et al. 2006).

Scattered trees could be used extensively to complement reserves and ecological networks by blurring habitat boundaries and creating additional secondary habitat for some species that cannot live exclusively outside reserves (see above). However, scattered trees would not, and should not, be a substitute for the other adaptation approaches, because not all species will be able to use such landscapes.

3.Biological legacies and expanding future options. Scattered trees can act as ‘biological legacies’ (Elmqvist et al. 2001) by acting as foci for tree regeneration, providing structural heterogeneity, enriching soils with water and nutrients for other organisms and ‘life boating’ of other species (Lindenmayer & Franklin 2002). In particular, the role as nuclei for restoration over wide areas could be critical through providing seed directly, or indirectly through seed deposited by perching birds and bats (Elmqvist et al. 2001). Embracing natural regenerative processes also has the advantage of creating scattered tree ecosystems that are self-organizing, resulting in structurally and functionally heterogeneous landscapes which could benefit biodiversity. Seed dispersal and availability of nurse shrubs also may be important for the long-term sustainability of scattered trees in some situations (e.g. Herrera & García 2009).

4.Integration of conservation and production. In creating more continuous cover in off-reserve areas, it is important to maintain production outputs. Scattered trees offer great potential to integrate conservation and production while facilitating adaptation. Scattered trees can be used in conjunction with agriculture without compromising production values. They also can have production benefits (see above). In regions where temperatures are predicted to increase, and rainfall to become more variable (e.g. south-eastern Australia, McAlpine et al. 2007), these values could become particularly important for both agricultural production and adaptive responses of biota.

Agroforestry methods also could be employed to maintain or establish scattered trees across landscapes in some situations. In particular, scattered trees lend themselves to grazing systems and could be encouraged with techniques such as rotational grazing (Manning et al. 2009). However, in some circumstance this may not be compatible with even low intensity management practices (e.g. Plieninger et al. 2003). In such situations temporal abandonment (i.e. rotational set-aside) to allow shrub encroachment of tree stands, before being re-opened for production, may be necessary to allow long-term persistence of scattered trees (Ramírez & Díaz 2008). This approach allows seed dispersal by animal vectors and facilitates early seedling survival where these processes are important (Ramírez & Díaz 2008; Herrera & García 2009).

The functioning of scattered trees as biological legacies also means that they can expand future restoration options in a changing environment. For example, Rey Benayas et al. (2008) propose the creation of small ‘woodland islets’ extensively across agricultural landscapes. Then, if land is abandoned, islets can act as sources of seed and woodland organisms. Scattered trees can serve a similar function. In scattered tree landscapes, land use could vary through time while trees persist, and could be used as ‘dynamic’ reserves (Bengtsson et al. 2003; Manning et al. 2009).

Future directions

Scattered trees have significant potential in facilitating adaptation to climate change. While the maintenance or creation of scattered tree environments will not be appropriate, feasible or desirable in all landscapes or situations, evidence supports their consideration as part of the mix of strategies in formerly wooded and forested landscapes. The list of possible ways that scattered trees could facilitate this adaptation outlined here is preliminary, and they could have many more values.

Despite their potential, scattered trees are threatened in many places under current management regimes. Ensuring appropriate levels of tree regeneration, and preventing premature mortality of mature trees, is critical for maintaining scattered trees in landscapes (Gibbons et al. 2008), and establishing new trees as part of any strategy . The encouragement of scattered trees would require the protection and restoration of traditional cultural landscapes, or the establishment of new ones.

Techniques for encouraging the regeneration and maintenance of scattered trees in agricultural landscapes might include planting, direct seeding or encouragement of natural tree regeneration using appropriate grazing regimes or rotational set-aside through time. Policy makers might encourage such actions through incentives, and should consider the protection and establishment of scattered trees in regulations and legislation.

The use of scattered trees by land managers might vary depending on the landscape or project, but is likely to be particularly important where conservation and production are being integrated. For example, ecological networks might be embedded within a landscape of scattered trees used for agriculture. Alternatively, scattered trees and integrated land uses could be nested within multi-use ecological networks.

The provision of continuous sparse cover allowing multi-directional connectivity, the potential to integrate conservation and production, the ability to complement reserves and ecological networks and to provide biological legacies and expand future land use options indicates that scattered trees could have a critical role to play in strategies to facilitate adaptive responses of organisms to climate change.


Thanks to Mario Díaz, Joern Fischer, Toby Gardner and an anonymous reviewer for comments and suggestions based on an earlier draft.