Trees have now been identified as one of the most significant groups of invasive plants (e.g. Richardson 1998). However, even though there are several well-documented cases of invasions by exotic forestry trees, these cases form only a small proportion of all the records in the literature. It should also be emphasized that not all of the effects recorded for invasive trees are detrimental: some human communities derive benefits from the invasive character of some species.
The main conclusions of this review and some recommendations are summarized below. The latter are shown in bold type.
1. A number of definitions of “invasive species” have been suggested in the literature, including some specifically for invasive trees or woody plants. However, none has been universally adopted by scientists or those working in applied fields such as agriculture and forestry. Some of the definitions distinguish between all invasive species and alien invasive species (the definition of “alien” is provided in Section 1). In terms of forestry and agroforestry, it may be more practical to confine the issues of invasiveness to introduced (i.e. alien) species as it is the scale of species introductions as a whole that is causing most concern on a global basis. Also, some communities associate invasive species with benefits, so invasiveness should be defined only as a biological character.
Recommendation 1. In the context of forestry, a common definition should be developed that focuses only on parameters of population expansion, since some definitions of invasive species imply only negative impacts, which is not always the case.
It is suggested that the definition provided by Cronk and Fuller (1995) has the most practical application in forestry:
Invasive plant – an alien plant spreading naturally (without direct assistance from people) in natural or semi-natural habitats, to produce a significant change in terms of community composition, structure or ecosystem processes.
2. On a global basis, only patchy information is available on the status of exotic forestry trees that have become invasive. The terminology used by authors is also very variable and there is frequent overlap in the terms “invasive” and “naturalized”. The evaluation of the extent of invasions by forest trees is generally very qualitative and subjective, making it difficult to assess the overall magnitude of the problem.
With these caveats in mind, the following summarizes the global situation. A total of 443 species, including 282 trees and woody shrubs used in forestry, 203 used in agroforestry and 292 amenity species, have been described as invasive (bearing in mind that some of these species are included in more than one usage category). However, there is little documentation that sets invasions within a wider context, despite the fact that, for each species, the scale of planting, the method and time span of introduction, establishment and management are all likely to vary according to their use, with consequent effects on their potential to become invasive.
Invasive species occur in many forestry genera, and are recorded in all parts of the world. However, there are certain regions where a potential information gap has been identified, e.g. Asia, northern and central Africa and parts of South America. Many species were reported as invasive in only a fraction of the countries in which they have been introduced, but the underlying reasons for this are not yet fully understood.
Recommendation 2. There is a need for further research and monitoring that will provide information on the management processes in planted systems and take account of the scale (i.e. land area) of plantings and of the area occupied by invasive species. Identification and monitoring of invasive species should be particularly supported in those areas where there is currently little documentation. Studies that focus on species which are invasive in some areas, but not in others, would be of particular value.
3. Few studies have been conducted on the positive and negative impacts of invasive forest trees. Positive impacts include fuel and other resources for resource-poor communities, as well as soil stabilization in overexploited natural forest areas. On the other hand, problems could potentially arise if introduced species hybridized to produce new invasive species; however, only a few such cases have been reported in forestry species (i.e. hybrids of Leucaena and Prosopis) and these are only locally important. Invasive forest trees have been reported as major problems in grassland pastures throughout the world but there are few instances where trees have invaded agricultural systems or forest plantations. Most reports of invasiveness relate to natural or semi-natural habitats. The latter include open forest systems, grasslands, riparian areas, wetlands and fire-dominated ecosystems. Some countries have quantified impacts on local species diversity and available ground water, but overall there is a severe paucity of information.
Although there has been a growing national and international awareness of the possible risks of invasive forestry trees, it is likely that some stakeholders in forestry remain ignorant of these risks. Awareness is highest in environmental sectors, but some of the risks have been highlighted by those in agricultural sectors. In some parts of the world, this has led to conflicts of interest that are fuelled partly by the general lack of quantitative information on the ecological and economic impacts of forestry trees. Such conflicts are compounded by a general lack of information on suitable tools (methodologies etc.) for making such assessments.
Recommendation 3. In light of the above, a number of case studies should be conducted in collaboration with countries that have a high degree of dependence on forestry. Such case studies should cover a range of forestry situations (commercial, developmental and environmental) and include the development and promotion of tools for ecological and economic impact assessments. Particular attention should be paid to those regions of the world where there is little information on the invasiveness of exotic forestry trees (e.g. tropical and temperate Asia).
4. Taken alone, many biological attributes (e.g. life history, taxonomic status and genetic constitution) are poor indicators of invasiveness. However, some of these characters (in combination with other factors such as the extent of invasiveness expressed), are being used in risk assessments (see below).
Recommendation 4. Rather than focusing on biological attributes, it would be of more practical value to examine those species that have been recorded as highly invasive in at least one area, in order to determine whether or not some form of control or management by local communities contributes to noninvasiveness in other areas. This approach could be combined with the studies suggested in Recommendations 2 and 3.
5. Globally, the development and implementation of prevention, control and management tools for invasive forestry trees has been cautious and patchy because of the various economic and developmental benefits of the trees concerned. Some countries, sometimes in collaboration with international partners, have made large investments in exotic trees. Thus, given the general lack of quantitative information on negative impacts, little action has been taken by many countries. The general lack of management tools (or information concerning existing tools) is an additional constraint that prevents many countries from implementing risk assessments, control and management schemes.
Work on prevention has included the development of risk assessment and risk management models. Practical risk assessments, based on information such as whether the plant is invasive elsewhere, are now in use in Australia, New Zealand and the USA. A few other countries/territories have schemes under development. The most common assessment methods to date are based on numerical scores (see Pheloung 2001).
Recommendation 5. There is an urgent need for methods of risk assessment to be evaluated further for use in forestry and, if found to be effective, promoted. These could also be incorporated into more general decision support systems that include socio-economic factors as well as biological risk.
6. Risk assessment and management of alien plants has also been considered in some cases. Many researchers in this field have called for monitoring schemes to be set up once a plant has been introduced. For forestry trees, this would involve planting trials and would need to be continued for many years.
Recommendation 6. Further work is required to develop practical guidelines specifically for monitoring forestry species, since no such guidelines are currently available.
7. In some countries, large eradication programmes employing mechanical and chemical methods have been undertaken against woody legumes (such as Prosopis) that are invasive in pastoral systems. However, experience has shown that these methods are costly and usually have not eradicated the trees concerned (although some success has been achieved in conservation areas in several countries (e.g. Mauritius and South Africa).
Recommendation 7. The small scale eradication project in Mauritius could usefully be used as a model for other areas.
8. Biological control and integrated control (biological with mechanical and chemical methods) have been used for the control of woody legumes in Australia, Southeast Asia and South Africa. These programmes are still largely ongoing and the results to date suggest that a combination of complementary natural enemies will be required to effect full control.
Some efforts are now being made to resolve conflicts of interest through the development of management (rather than control) programmes for invasive trees. In South Africa, on the basis of economic models, seed feeding bruchids have been introduced to control the seed output of several legume trees that have become invasive (e.g. Acacia mearnsii) and these efforts are reported as being successful. These programmes have been supported by new legislation that restricts the planting of trees that have invasive tendencies. At more of a research level, work in pasturelands in other countries has shown how, under some management regimes, invasive woody legumes can complement pasture grasses for livestock feeds.
Recommendation 8. All these various experiences in management should be used as models for other countries trying to resolve issues associated with invasive forestry trees.