This section is based largely on Wittenberg and Cock (2001).
Prevention is the first and most cost-effective line of defence against invasive alien species. For unplanned introductions, the benefit/cost ratio for exclusion is undoubtedly high. Even if only a small fraction of inadvertently introduced species become established and only a small fraction of those become invasive pests, the enormous cost of some of the latter can far outweigh any unexpected benefits that unplanned introductions may confer. For example, in the USA, the estimated cumulative potential losses from the Asian strain of the gypsy moth (Lymantria dispar) and the nun moth (L. monacha) between 1990 and 2004 are in the range of US$35 billion to US$58 billion (net present value in 1991 dollars). Thus, while the ongoing cost of preventing the introduction of the moths and of treating potentially contaminated goods is not negligible, it is far lower than the potential losses resulting from their establishment (OTA 1993).
The costs of establishing and maintaining the exclusion apparatus can be quite high, although mechanisms exist for apportioning fees to the users and beneficiaries of the system. For example, in New Zealand, private parties pay all costs associated with risk analysis and port inspection for imported alien species. In contrast, in the USA, while the commercial interests advocating Siberian timber imports (Case Study 22) spent about US$200 000 on developing Russian contacts and promoting the products, the US Government spent approximately US$500 000 more on analysing the associated risks (OTA 1993).
On average, it has been estimated that approximately 1 percent of introduced species are likely to become invasive (Williamson 1996); this estimate, coupled with the fact that eradication or control of well-established species is both difficult and costly, suggests that the precautionary principle should be adopted, i.e. that all species should be treated as potentially harmful until the contrary is proven (Case Study 10).
Targeting individual species is the most common approach to preventing the introduction of invasive organisms. However, a more comprehensive strategy is to identify the major pathways through which harmful invasions occur and then manage the associated risks. Pathways as well as individual species can be subjected to risk assessments, and exclusion methods based on the latter would probably be more efficient, since efforts could then be concentrated where pests are most likely to enter. At present, while international trade and travel are believed to be the leading cause of harmful unintentional introductions, in most countries there is a lack of solid evidence concerning the actual pathways. In the case of forestry pests, for example, forestry products such as packaging materials (Case Study 20) can be particularly important in facilitating the worldwide movement of pest species.
It has been argued that, because some pathways have been used for decades (or even centuries) without any prevention methods, all invasive species have probably already spread to most areas. However, it is now obvious that establishment rates can vary over time, with some alien species known to have been introduced to a particular area decades ago only recently becoming established. Reasons for this delayed establishment include alterations of the alien species itself, changes in the pathway, climatic changes and changes in human impact in the area of introduction (e.g. growing forests near entry points). The accelerating rate of establishment of alien species demonstrates that the concern over accidental introductions is still valid. Furthermore, new pathways will be created with increased mobility and trade.
One such example, relevant to forestry, is the international trade in bonsai trees. China is now a major source of bonsai trees, exporting nursery stock to many countries. Pests can accidentally be transferred on the plants or in the soil. For example, the citrus longhorn beetle, Anoplophora chinensis (Forster) was introduced from China to Italy on bonsai and the USA quarantine service often intercepts beetles on bonsai nursery stocks (American Lands Alliance 2002; USDA-APHIS 2002)
The responsibility for preventing the introduction of alien species is normally a function of the government agricultural administration rather than the forestry sector itself. Thus, while the latter should be an important partner in these efforts (particularly in contributing to pest risk assessments), prevention should be managed, prioritized and implemented on a broader scale to address all risks, not just those posed by the priority pathways. For this reason, prevention is not considered further here, except inasmuch as forestry activities may provide a pathway for the introduction of alien species (Section 6.3).
The longer an alien species goes undetected once introduced to a new country, the less opportunity there will be to intervene if it becomes a problem: the options for its control will become fewer and any intervention will become more expensive. For example, eradication (see Section 3.3) will rapidly cease to be an option once an alien species is left to reproduce and disperse. The opportunity for eradication or early control of a new colonizer therefore makes investment in detection worthwhile. More specifically, priority should be given to detecting and monitoring those species known to be invasive elsewhere, especially those spreading within a region.
Surveys aimed at detecting new invasive species should be carefully designed and targeted to answer specific questions as economically as possible (e.g. Case Study 4). Usually all that is needed from such surveys is an indication of presence or absence. However, some invasive species are more easily detected than others: cryptic species may require special efforts to locate or identify, particularly when present in low numbers. For certain groups of pests, surveys by experts should be made to permit a rapid response before the species becomes well established. For a detailed treatment of this subject see IPPC (1997).
The biosecurity surveillance programmes in New Zealand provide a useful model: although they specifically target invertebrate pests and diseases, there is also a generic surveillance programme of "high risk" sites (e.g. ports of entry, unpacking facilities, car import yards and industrial premises where machinery is unloaded). Staff from the Biosecurity Ministry (or contractors) also visit parks and other urban forest stands and monitor for symptoms of invertebrate pests or diseases. Public involvement is also encouraged and as a result some newly established exotics have been detected at an early stage (Case Study 11). Surveys are undertaken in a "walk through" manner and samples are taken where signs of ill health or feeding damage are observed. In this way, several plant pathogens are detected each year. Some are studied further, e.g. pine pitch canker, but in general the MAF response will be to follow the guidelines stated in their biosecurity policy (MAF Biosecurity Authority 2001).
Many pest species become invasive only after a considerable lag time during which they persist in small numbers until outbreaks occur. Several causes for this delay are discussed in the literature (e.g. Crooks and Soule 1996; Williamson 1996; Kowarik 2003), including alteration to ecosystems caused by land use changes, or changes in agricultural and forestry practices which may favour some species over others, or create new pathways linking habitats, etc. Other suggested causes include a process of adaptation to the new habitat, perhaps through a genetic bottleneck. Some species may simply be going through a long period of exponential growth and may therefore appear to suddenly increase in population size and become invasive. Since this type of delay is most likely to be associated with long generation times, many species of exotic trees planted around the world are likely to be "sleeper" invasive alien species, a fact which will only become apparent in years to come.
Contingency plans should be prepared to deal with any alien species found during surveys. Such plans are usually a carefully considered outline of the action that should be taken when a new invasive species is found. However, given the diversity of potentially invasive alien species, and the variety of options for their control, plans will initially have to be either very broad - perhaps identifying general principles, responsibilities and likely stakeholders - or targeted towards specific high risk species or groups. Over time, more specific components for different groups or species may be added to the overall plan to provide a detailed contingency plan for more general use. Of equal importance to formulating the contingency plan is securing the involvement and commitment of all those involved in caring for the area at risk. Not only must they all understand the plan, they must also put into immediate effect those parts of it relating to the prevention and early detection of alien species.
When preventative methods fail, an eradication programme is usually the preferred method of action for species that pose a significant risk. Eradication is the elimination of the entire population of an alien species, including any resting stages, within the managed area. Where it is feasible, eradication is often a successful and cost-effective solution. However, because eradication programmes often appeal to politicians and the public, there is often a temptation to implement them even if they are unlikely to succeed. In reality, containment may be a more realistic alternative (Case Study 12). A careful but rapid analysis of the risk posed by a new species, the probable costs (including indirect costs) and the likelihood of success must be made; even then, eradication should only be pursued when the necessary funding and the commitment of all stakeholders (including the public) have been secured. Eradication of mammals, particularly those with which humans can identify, is particularly prone to opposition. Methods of killing these targets are rightly the subject of discussion and are often the cause of disagreement. However, resolving such conflicts takes time, and eradication is more likely to succeed if the initial response is rapid. Decision-making in such cases is not easy and must take these conflicting viewpoints into account.
The best chances of successful eradication are during the early phase of invasion, while the target populations are small and/or limited to a small area; well-established populations and large areas of infestation are usually unsuitable for eradication programmes. Furthermore, once an alien species is well-established and has become a major element of the ecosystem, its eradication (or control) may influence that ecosystem in unpredictable ways. The relationships between the target species and others (both indigenous and alien) then have to be carefully considered. Elimination of a dominant alien plant may lead to its replacement by one or more other alien species, perhaps with more damaging characteristics.
Eradication efforts have been most successful in island situations, including `ecological islands' isolated by physical or ecological barriers, e.g. forest remnants surrounded by agricultural fields. However, the target species may survive in small populations outside an ecological island and, depending on the degree of isolation, could rapidly reinvade after an eradication campaign. The same may be true of real islands, especially coastal islands and archipelagos.
Eradication programmes can involve several control methods, either alone or in combination. The methods used vary with different species, but may include:
Some groups of organisms are more suitable for eradication efforts than others. Each situation should be carefully evaluated to determine the best method under the given circumstances. Ideally, the methods chosen should be as specific as possible, but the rather rigorous nature of concentrated eradication efforts will often inflict incidental casualties on nontarget species. In most cases, however, these losses are acceptable when balanced against the long-term economic and biodiversity benefits.
In general, plants are best eradicated by a combination of mechanical and chemical treatments, e.g. cutting woody weeds and applying herbicide to the cut stems. Amongst land invertebrates, only some snails and insects have been successfully eradicated (Case Study 11). Snails can be hand-picked, whereas the commonest options for eradicating insects are insecticides or biopesticides, usually by aerial application and/or in the form of baits. Mass releases of sterile males have been effective on several occasions (e.g. for fruit flies and the screw-worm fly) but have not been used in forest systems.
The advantage of eradication as opposed to long-term control is the possibility of complete reversion to the conditions prevailing prior to the invasion of the alien species. There are no long-term control costs (although precautionary monitoring may be appropriate) and the ecological impacts and economic losses fall to zero immediately after eradication.
The major drawback of eradication programmes is that they may not succeed, in which case the entire investment is largely wasted - at best, the spread of the target species will have been slowed. Many failed attempts have been both costly and have had significant adverse effects on nontarget species (e.g. the attempt to eradicate South American fire ants in the southern states of the USA, which ultimately had to be abandoned).
In summary, the basic criteria for a successful eradication programme are as follows:
Species currently considered unsuitable for eradication may be targeted in the future as new technologies are developed. For example, genetically engineered micro-organisms are currently being evaluated for the eradication or control of introduced foxes and rabbits in Australia. Such approaches may become more widespread in the future if public concerns over their safety can be satisfactorily addressed.
Once an alien species has become established and eradication is no longer an option, a particularly useful management tool is biological control by the introduction of exotic natural enemies from the pest's area of origin (e.g. Case Studies 13 and 14). This approach (so-called `classical' biological control) is particularly appropriate for the control of alien pests and has been carried out successfully for more than a century (Greathead and Greathead 1992; Julien and Griffiths 1998).
Despite the undoubted success of some biological control programmes, practitioners have become increasingly aware that introduced biological control agents may themselves have undesirable side effects. Initially, this concern was limited to the possible impact of introduced biological control agents on economically important plants and insects (notably honey bees and weed biological control agents). More recently, increased attention has been given to the potential danger to all indigenous fauna and flora, particularly rare and endangered species.
All biological control projects should therefore have a sound scientific basis and be subjected to proper risk assessments, with all stakeholders being involved in the decision-making process. Nevertheless, it should be remembered that any introduction is a permanent decision and that a successful biological control agent may sometimes spread to unanticipated areas (Case Study 15). Biological control agents are normally quarantined upon importation to screen for parasites and diseases, and to check the purity of the material.
In response to these concerns, some countries (mostly in the developed world) have developed their own legislation, whereas others depend on the IPPC (1996) Code of Conduct for the Import and Release of Exotic Biological Control Agents. The latter was an attempt to formalize current good practice and to set standards in terms of the information needed for decision-making without being over-restrictive (Kairo et al. 2003).
There are various other management options for the control of insects, diseases and weeds, although many are unsuitable for use in forest systems on the grounds of practicality, effectiveness, economics or environmental risk. Ideally, all relevant and appropriate techniques should be integrated to meet the pest management objective in an optimum manner (e.g. maximizing yield and profit while minimizing adverse environmental and aesthetic impacts). Suitable approaches include cultural techniques (e.g. seedling management, planting patterns, buffer zones (Case Study 16), ecoclimatic matching, resistance/ tolerance to pests), mechanical methods (e.g. cutting, bulldozing, shading), chemical control (pesticides), and biological control (classical, augmentative, or conservation). The principles and use of integrated pest management (IPM) are considered beyond the scope of this report, although an overview is presented in Table 1 and Figure 1. For more information, see relevant texts from forestry (e.g. Speight and Wylie 2001) and agriculture (e.g. Dent 2000; Matthews 1984; Mengech et al. 1995; Metcalf and Luckman 1994; Morse and Buhler 1997; Norton and Mumford 1993).
|
Case Study 16: Buffers to keep invasive plants out of forest fragments Fragmented forest habitats are vulnerable because they have proportionately more edges through which alien species can invade. The obvious solution of minimizing the amount of edge is not always feasible, but there may be another effective approach: intact edges can help keep seeds out of the forest interior. Cadenasso and Pickett (2001) quantified aspects of this phenomenon in New England, USA. They compared the number of seeds blowing from an old field into an adjacent deciduous forest patch through two types of edge: intact and thinned. The thinned edge was created by removing all trees, shrubs and branches that were less than half the height of the forest canopy. This thinned edge extended 20 m into the forest patch and resembled that created by logging or a large blow-down. Four times as many wind-borne seeds crossed the thinned edge than the intact edge, and seeds penetrated 2.5 times further into the thinned forest edge than the intact edge - 45 m vs 17 m. Cadenasso and Pickett therefore recommended protecting forest fragments from invasive weeds by "sealing" the edges - removing non-native plants and planting with dense native shrubs, vines and understorey trees. A similar result was obtained in New Zealand, where a buffer of Pinus radiata (planted for timber production) effectively protected an adjacent native podocarp-broadleaf forest from invasion by exotic species. The vegetation at the interface of the two zones was a mixture of light and shade-tolerant species, reflecting its history (i.e. an 'open' edge both prior to the planting of the P. radiata and during its early growth phase, and a 'sealed' edge once the pine reached a certain height and density). Very few exotic plants were found in the native forest adjacent to the mature pine, although such species were found both under the pine and within the first 10 m of the native forest where it was adjacent to pasture. The edge of the native forest area adjacent to 75-year-old pines (with diameters up to 70 cm) was indistinguishable in species composition from the interior of that area of forest. At the edges of a stand of 25-30 year old pines, the vegetation was similar in composition to that found 25 m into native forest adjacent to pasture (Karen Denyer, unpublished report). In addition to reducing the spread of invasive alien plants (especially those with wind-borne seeds), such buffer zones can provide new habitats and increase local species diversity. They can also limit other adverse effects from adjacent land. Specific benefits include reducing fire risk, reducing the risk of pesticide spray damage, protecting forest edges from wind penetration, protecting sensitive plants and animals in the interior, and limiting the movement of sediments and nutrients (particularly in areas adjacent to wetlands and riparian habitats). Sources: Cadenasso and Pickett (2001); Davis and Muerk (2001); Denyer (2000); http://www.canr.uconn.edu/ces/forest/invasives.htm |
Table 1 : Components of a generalized IPM system.
Stages A and B are entirely preventative, Stage C involves monitoring and prediction, while Stage D covers control strategies should prevention fail or monitoring reveal high risk (adapted from Wylie 2001; Speight et al. 1999).
Stage |
Options |
A |
|
B |
|
C |
|
D |
|
Control techniques used in plant nurseries are likely to be more flexible and more intensive than techniques available for growing plantations. For example, seed treatments, fertilizer applications, irrigation treatments, drenches, hand-picking of pests, insecticide applications, etc. may all be used to good effect in nurseries while being unsuitable for maturing plantations.
An underlying tenet of pest management is that a healthy crop will minimize pest damage and its impact. This is particularly relevant to forestry systems, where interventions for pest management should be minimized or eliminated after planting out. Seedlings which have been protected from pests and which are planted with a nutrient-rich planting substrate are more likely to survive new pest attacks than material that has been stressed before it is even planted. Hence investment in protection at the nursery stage helps prevent attack by both forest pests and diseases, particularly in the case of individually planted agroforestry trees.
It is important to remember, however, that the key pests are likely to be different at different stages of tree growth, and that many pests will attack maturing trees, regardless of whether the seedlings were healthy or not. Hence plantation managers will begin implementing their pest management strategies well before they start putting seedlings in the ground (Wylie 2001). Choosing species and provenances that are resistant to potential pests and which suit the site (and which will therefore grow vigorously) is clearly important. However, resistance can be overcome if selection pressure is high enough, and planting a mixture of resistant varieties - if more than one is available - is probably the best strategy. Multi-species (or even single-species) plantations with a 'mosaic' of tree ages are also less likely to experience catastrophic pest infestations than even-aged monocultures.
Figure 1 : Summary of factors that may interact to create insect pest outbreaks in tropical forestry and strategies for reducing risk
(from Wylie 2001; Speight 1997)
In agricultural crops, pest management decisions are often based on the use of economic thresholds, which take into account the revenue losses resulting from pest damage and the costs of treatment. Below the economic threshold, the presence of the pest is tolerated, and only when damage rises (or is predicted to rise) above the threshold is action taken. However, this approach is difficult to apply in forestry: determining the economic threshold for long-lived perennial crops like trees is particularly difficult because economic and biological forecasts must sometimes be made over decades (Wylie 2001).
Furthermore, IPM relies heavily on monitoring to identify areas where pest populations are high and when economic thresholds are likely to be exceeded (Clarke 1995). In contrast to intensive agriculture, however, such monitoring may be impractical or at least inaccurate in large and inaccessible forest areas. Adopting an IPM approach can be complex and expensive when detailed monitoring is required. This is obviously a constraint in many tropical forestry operations where profit margins are often low.
The Forest Stewardship Council (Case Study 17) has formulated principles for environmentally appropriate, socially beneficial and economically viable management of the world's forests. These principles recognize many of the ecological issues raised here, including the risk of introduced tree species becoming invasive. However, they also recommend minimizing the use of biological control agents and not using genetically modified organisms - statements with which not everyone would agree.
Other, similar schemes (e.g. the Montreal Process (1999), and the Sustainable Forestry Initiative (SFI 2002)) pay less attention to biosecurity, or address the management of natural forests, where some of these issues are less relevant.
There are some discernable trends in national approaches to forest pest management, notably towards more sustainable management strategies. These changes are often related to a changed perception of the role of forests, which are now seen not only as a source of commercial products, but also as valued ecosystems in their own right, with associated ecological and social functions.
As far as pest management is concerned, the main tendencies are as follows:
· Pest outbreaks are no longer automatically viewed as a calamity, particularly in forests where logging is not the main function. As a result, the use of pesticides in forestry has greatly declined. Some countries, such as Switzerland and Sweden, have banned, or are in the process of banning, the use of pesticides on forest trees (e.g. Zahn 1999). In Canada, insecticide use in forestry has been severely restricted in recent years: registering pesticides for use in forestry has become increasingly difficult and most provinces allow only microbial insecticides (mainly Bt) in the natural forest environment (Carrow 1995).
· In the development of new, sustainable pest management methods, increasing emphasis is being placed on prevention rather than control. In Europe, forest health monitoring networks were set up at both national and EU levels during the 1980s and 1990s, partly as a result of concern over a possible large-scale decline in forestry (e.g. DSF 2000). To help coordinate surveying, monitoring and prevention methods on a European scale, a IUFRO working party was created in 1998 and has met yearly ever since (Forster et al. 1999).
At the moment, these trends are most clearly seen in developed countries, but in future are likely to be more widely adopted in developing countries as well.
2 The following discussion of eradication options and criteria is based substantially on Wittenberg and Cock (2001).
