Previous PageTable Of ContentsNext Page

Forest biodiversity at the ecosystem level: where do people fit in?

J.A. McNeely

Jeffrey A. McNeely is Chief
Scientist at the World Conservation
Union (IUCN), Gland, Switzerland.

Diversity is needed in approaches to managing forest ecosystems, just as in the forest ecosystems themselves.

It is now widely accepted that biodiversity is the measure of biological variety at many scales, from the gene to the ecosystem. Much attention has been given to diversity of species within forests, especially trees that are being harvested. The challenge is to see the forest for the trees: How can the idea of biodiversity at the ecosystem level be translated into practical action for better management of forest ecosystems?

The definition of "ecosystem" is reasonably well agreed. For example, the Convention on Biological Diversity (CBD) defines the term as "a dynamic complex of plant, animal and micro-organism communities and their non-living environment interacting as a functional unit". The living components of an ecosystem interact in very complex food webs (Schoener, 1989). Ecosystem approaches to forest management take into consideration the complexity of these interactions and seek both to maintain the productivity of forest ecosystems and to enhance their capacity to adapt to change.

Focusing on the ecosystem level provides a strong basis for solving critical problems in resource management. For example, conserving forest biodiversity at the ecosystem level helps to support services such as maintaining the balance of atmospheric gases, recycling nutrients, regulating climate, maintaining hydrological cycles and creating soil (Daily, 1997). While scientists are still developing their understanding of the relationships among taxonomic diversity, productivity, stability and adaptability of ecosystems, new research indicates that species diversity enhances the productive capacity of many forest ecosystems and their ability to adapt to changing conditions (Johnson et al., 1996).

Another important implication of considering forest biodiversity at the ecosystem level is the potential of mismanagement leading to the essentially permanent transformation of a highly productive forest into a much less productive system (such as a grassland). Recent research has indicated that even gradual changes in climate, the flow of nutrients, extraction of natural resources and habitat fragmentation can lead to sudden drastic switches in the character of a forest ecosystem (Scheffer et al., 2001). While many different factors can lead to such shifts, a critical factor is a loss of resilience (the ability to recover from external events) through declining biodiversity at the ecosystem level.

In seeking to apply ecosystem approaches to forest biodiversity, especially forests that are being influenced by increasing levels of use by people, it is helpful to focus on some key questions. First, are people part of forest ecosystems? Second, what are the impacts of human harvesting on forest ecosystems? Third, how can forest ecosystems be managed so that they provide both the goods and the services that are required by modern society? This article briefly explores these key issues, and indicates appropriate lines of action to be taken in future.


The ecological literature is replete with terms such as "primary forest", "undisturbed forest" and even "primeval forest". But a growing body of evidence indicates that virtually all forests on the planet have been substantially influenced by humans, most for at least several thousand years. Studies by foresters, ecologists, historians and anthropologists on forests in tropical, temperate and boreal regions conclude that forests and people have evolved together over thousands of years, with people planting the trees they prefer, using fire to burn forests to improve hunting conditions, and managing forest fallows to maintain their agricultural fields. Far from being "primeval", forests are part of the human landscape, and the biodiversity found in today's forests has been profoundly influenced by people.

For example, before the voyages of Christopher Columbus brought North American resources to the attention of Europe, the people living in the eastern woodlands of the United States were a "potent if not crucial ecological factor in the distribution and composition of the forest" (Williams, 1989). When the indigenous people were extirpated from the woodland-grassland zone and permanent European settlements replaced them, the woodland became much more widespread, and the density of the woodland increased with the intensity of the development. In essence, both native peoples and the European colonists designed the forests they preferred.

The vast boreal forest-covered wilderness of northern North America is often considered to be natural. But people have occupied this forest from its very beginnings, as the great ice sheets withdrew northwards at the end of the Pleistocene era. New studies have established that native Americans in northern Alberta, Canada, regularly and systematically burned habitats to influence the local distribution and relative abundance of plant and animal resources. This pyrotechnology is similar to that reported for hunter-gatherers elsewhere in the world, which created an overall fire mosaic that characterizes the northern boreal forests (Lewis and Ferguson, 1988). Hunter-gatherers elsewhere in North America and in several parts of Australia also employed habitat fires, specifically in the maintenance of "fireyards" and "fire corridors" in widely separated and different kinds of biological zones.

Further south, Gomez-Pompa and Kaus (1992) found that many of the tree species now dominant in the mature vegetation of tropical America are the same species protected, spared or planted when the land was cleared for crops as part of the ancient practice of shifting agriculture. They contend that the current composition of mature vegetation is the legacy of past civilizations, the heritage of cultivated fields and managed forests abandoned hundreds of years ago.

Numerous other examples can be cited from Amazonia (Roosevelt, 1994), Central Africa (Fairhead and Leach, 1998), Europe (Delacourt, 1987) and tropical Asia (Spencer, 1966), but the conclusion is clear: While forest ecosystems are "natural", humans are an essential part of this "nature". Hence building resilience into forest ecosystems requires building resilience into the human management systems, enabling them to adapt to changing conditions.

Most forests have been influenced by humans, and forests and people have evolved together over thousands of years - for example, many forests in Sumatra, Indonesia have been cleared repeatedly by shifting cultivators for millennia, yet the forest still provides rich resources on which the farming inhabitants of the forest edge depend



While the human impact on forest ecosystems has been profound throughout history, only in the past few decades has the human influence spread comprehensively and simultaneously in virtually all forests. By far the greatest impact has been in forest clearing, both to create new agricultural lands and to harvest valuable timber. A review of a wide range of studies on the impacts of logging practices on tropical forest ecosystems and biodiversity (Johns, 1997) indicated that logging of mature forest commonly leads to a local increase in species diversity as structural and associated microclimatic changes create patches of habitat and food resources attractive to species that typically live in secondary forest and on forest edges. However, populations of many species that typically live in the forest understoreys markedly decline and remain locally low or absent for many years. Johns (1997) concludes that the most appropriate way to manage tropical forests for producing timber, without losing other values, is to have small undisturbed forest areas preserved within a larger matrix of production forest, a prescription that is being attempted in some parts of Malaysia. On the other hand, commercial loggers have been notably reluctant to adopt sustainable forestry practices because they earn greater short-term profits when they externalize more of the costs, such as conservation of biodiversity.

In tropical forests, the large canopy and emergent trees that are most attractive to loggers are critically important sources of food (fruits and flowers) and shelter for animal populations. They are reproductively dominant and strongly influence forest structure, composition, gap dynamics, hydrology and biodiversity. Forest fragmentation in central Amazonia is having a disproportionately severe effect on large trees, the loss of which will have major impacts on forest ecosystems (Laurence, 1999). Mortality of trees, and especially of large trees, is highest near forest edges. This has significant implications for conserving rain forest ecosystems and the biodiversity they contain. The rapid rate of mortality of large trees may reduce the fecundity of canopy and emergent species, diminish forest volume and structural complexity, promote the proliferation of short-lived pioneer species and alter biogeochemical cycles affecting evapotranspiration, carbon cycling and greenhouse-gas emissions - key ecosystem services.

The problem is not simply the result of harvesting trees that support other species. In the Congo, roads established and maintained by logging concessions intensify bushmeat hunting by providing hunters greater access to relatively unexploited populations of forest wildlife and by lowering the costs of transporting bushmeat to market (Wilke et al., 2000). The trade in bushmeat is now reducing many species to mere remnants in many parts of the African forest zone. Reconciling the contrary effects of roads on economic development and biodiversity conservation is one of the key challenges to ecosystem managers in all nations. Failing to address this problem could lead to forest ecosystems that are virtually empty of wildlife populations that play essential roles in pollination, seed dispersal and nutrient cycling (Redford, 1992). This is no trivial matter; about 70 percent of the trees in the Atlantic forest of Brazil have seeds that are dispersed by vertebrates, mostly birds and mammals (Cardoso Da Silva and Tabarelli, 2000). Where key vertebrate dispersers have been eliminated, seed flow of tree species through the landscape is very limited, and the large fruit-producing species are being replaced by others that may be less useful. Such processes can lead to profound, and unpredictable, changes at the ecosystem level (Scheffer et al., 2001).

Deforestation is widely recognized as a major conservation issue, but the related issue of habitat fragmentation receives insufficient attention. In the Brazilian Amazon alone, the area of forest that is fragmented (with forests less than 10 000 ha in area) or prone to edge effects (less than one kilometre from clearings) is over 150 percent greater than the area that is deforested. A similar pattern is found throughout the tropics, so the fate of the world's tropical forest ecosystems is greatly affected by the capacity of their various species to survive in fragmented landscapes. Small fragments have very different ecosystem characteristics than larger areas of forest, supporting more light-loving species, more trees with wind- or water-dispersed seeds or fruits and relatively few understorey species (Laurence, 1999). As mammals and birds that disperse fruits disappear from these habitats, trees with fruits dispersed by them decline. The smaller fragments also have a greater density of tree falls, a more irregular canopy, more weedy species and unusually abundant vines, lianas and bamboos; thus they conserve only a subset of the original flora, and the fauna that is adapted to these species.

As human impacts on forests continue to increase, areas that were once continuous forest with sporadic clearings become agricultural landscapes with sporadic forests. This leads to a significant decline in population for at least some species of forest birds because fragmentation reduces nesting success, and thereby the number of offspring that can be produced. One recent study found that the reproduction rates were so low for some species in the most fragmented landscapes that their populations had to depend on immigration from other populations from habitats that had more extensive forest cover (Askins, 1995). Conservation strategies need to ensure the preservation and restoration of large, unfragmented forest habitats in each region (Robinson et al., 1995; Askins, 1995) and to support greater efforts to build linkages between ecosystems at the landscape level (Bennett and Wit, 2001).


Ecosystem approaches to conserving forest biodiversity should recognize that all environmental policy is best considered as the testing of a hypothesis, where proposed management actions are expected to target precise objectives and lead to predicted results. In this sense, ecosystem management is always an experiment, always an exercise in learning from experience. An essential element to feed learning back into ecosystem management is monitoring, which provides the information basis for modifying management actions in light of experience. Many ecological networks have now been established, indicating how such monitoring and feedback systems can function in a wide range of forest ecosystems (Bennett and Wit, 2001). An important aspect will therefore be to define the management objective as precisely as possible, taking into account the information available.

In seeking to manage forest ecosystems, it is important to keep in mind that resource managers are dealing with systems that are dynamic at many scales, ranging from individual leaves up to very large landscapes (Holling, 1992). The natural range of variability at each of these scales is often very wide, and it is not yet possible to predict how changes in the patterns and processes at any given scale are likely to affect processes at other scales. Deciding the extent and level of appropriate human impacts on such constantly changing systems is challenging in the face of limited knowledge. But new tools and techniques, such as remote-sensing imagery, simulation modelling, geographic information systems, and the increased capacity of data processing can help enrich understanding of the dynamism of forest ecosystems, and thus help enhance human capacity to adapt to changing conditions. Considering forest biodiversity at the ecosystem level helps to reinforce this perspective.

What can be done to conserve forest biodiversity at the ecosystem level? While this question remains a challenging one for resource managers, some general directions already seem apparent. First, protect large areas of forest rather than small ones where possible. Second, rebuild connectivity among small adjacent protected areas by including intervening habitat and promoting reforestation of the landscape. Third, protect forest edges against structural damage, damage by fire and colonization by exotics, by leaving a natural buffer zone of forest that could be managed to resemble a natural ecotone (a boundary or transition zone between adjacent communities) rather than an abrupt edge. Finally, minimize the harshness of the adjacent matrix by diversifying and promoting less intensive types of land use around forests, controlling the use of fire in ecosystems that are not fire climaxes (plant communities whose succession is maintained by periodic fires), minimizing the application of toxic chemicals and controlling the introduction of potentially invasive alien plant species. This approach is well illustrated by the proposed large-scale biodiversity corridors in Meso-America, Amazonia (Gascon, Williamson and da Fonseca, 2000) and elsewhere. The general aim is to apply the principles that enable forests to function as ecosystems to the practice of forest management, for example by ensuring natural regeneration, using low-impact logging that does not disrupt soils and avoiding excessive fragmentation.

Forest resources help support sustainable development in lowland areas of Sulawesi, Indonesia, suitable for rice cultivation; land-use plans can help decide the extent and level of appropriate human impacts on dynamic ecosystems


Given that people are part of forest ecosystems, involving local communities may be an important way to help resolve conflicting interests between local people and forest departments and may help contribute to conservation objectives. For example, in Nepal, management of village forests by local forest user groups has created a strong feeling of ownership and helped improve forest management practices. Local men and women, including poorer members of the village, are involved in activities like thinning, pruning, fire control surveillance and harvesting. As a result of their efforts, species composition of flora and fauna, crown coverage, habitat and micro-habitat of invertebrates, mosses, fungi and lichens are all improved, with positive impacts on the forest ecosystems. The community forests are providing ecological stability and the forest user groups are becoming more sensitive to conservation objectives. In at least some forests, wildlife populations have increased along with species diversity (aus der Beek, Rai and Shuler, 1997).

The living dead

The "living dead" are plant species that are still represented by living individuals but are incapable of reproducing because the animals on which their reproductive cycle depends have been eliminated from the ecosystem.

A well-known example is the calvaria tree (Sideroxylon majus) on the island of Mauritius, which ceased regenerating following the extinction of the dodo (Raphus cacullatus), because its seeds could not germinate without passing through the gizzard of a dodo. The tree species was saved when experiments demonstrated that the seeds would germinate if they passed through the gizzard of a domestic turkey.

While local and indigenous people are as tempted as anyone else to overexploit forest resources for short-term gains, some have instituted their own ecosystem management measures. For example, the Emberà, a group living in the forests on the Colombia-Venezuela border, reserve large areas in upper watersheds and along the spines of mountain chains as areas protected by spirits; the areas that benefit from this protection are remarkably similar to those typically set aside as protected areas by modern governments. These large areas of old-growth forest provide refuges for the reproduction of wildlife resources and protection of watersheds. The Emberà maintain ecological stability through a range of techniques that have parallels in many other parts of the world, both North and South: local technology; protection of important sites; appropriate settlement patterns; flexible social rules; egalitarian social structure; religious commitment; and a strong tradition of self-interested management of forest resources (Harp, 1994).

In the Arun Valley, Nepal, watershed forests help maintain the productivity of irrigated rice fields while still producing a flow of non-wood forest products; the trade-offs among resource users (farmers, foresters, fisherfolk and others) can best be addressed at the ecosystem level


The faith in local communities as forest ecosystem managers needs to be balanced with recognition that forests achieve numerous national objectives, including meeting needs for timber and fuelwood, retaining options for future economic use, addressing ethical and aesthetic values and providing global benefits such as biodiversity conservation. Thus simply ensuring local management of forest resources may not always lead to socially optimal levels of biodiversity conservation. Instead, the larger society must mobilize additional resources and approaches to support a socially desirable level of conservation effort, appropriate for its ecological, social, historical and political setting. As in every other field, management means setting objectives and making the trade-offs necessary to achieve them.


Previous PageTop Of PageNext Page