FAQs

Before any natural enemy is released, detailed host specificity testing is conducted to determine if other plant species are likely to be attacked. These tests may involve examining other plants in the native range of the natural enemy as well as laboratory-based choice (includes the target species) and no-choice (test plant only) trials. The results of all tests are taken into consideration before determining if a natural enemy is suitable for release. If there is an unacceptable risk that other desirable plants species will be attacked, as demonstrated in any of these trials, then the natural enemy is not released. Sometimes a decision is still made to release a natural enemy if the risk to other plants is deemed acceptable e.g. a closely-related introduced plant with no/limited beneficial uses that may also have potential to become a future weed.

No. The natural enemies have co-evolved and adapted to only attack their host plant(s). They are unable to recognise other plant species as hosts. Therefore, if the target weed disappears for any reason, the natural enemy will die out. However, in reality, the host plant rarely dies out completely and there are always a few plants to support the natural enemy.

No. See above. Natural enemies have co-evolved and adapted to only attack their host plant(s). They are unable to recognise other plant species as hosts. Therefore, if the target weed disappears for any reason, the natural enemy will die out.

In general, natural enemies rarely deplete their target weed to such an extent that there are no plants remaining. Normally, natural enemies reduce the target weed to low levels, such that there are always a few plants remaining and thus able to sustain low populations of the natural enemy.

Very rarely. There was one case in Australia many years ago where a natural enemy of lantana was found to attack a close relative of lantana. This close relative was an introduced garden plant species and was never tested. The insect is still found on the plant occasionally but does not kill the plant. In most situations, any natural enemy that attacks other plants was known to do so before it was approved for release. In these cases, either the damage by the natural enemy to that plant was found to be negligible or that the affected plant is not of economic or environmental concern.

No. Natural enemies utilized for biological control have co-evolved with their host plant and so are intricately linked to that host plant and do not recognize other plants as potential hosts. A beetle released on an exotic thistle in the USA in the 1960s has also attacked native thistles. Researchers knew this would happen but were not concerned as the damage was minor. To this day, the natural enemy has not caused problems and is not considered a pest of the native species.

Any non-target impacts on other species are picked up during the host specificity phase of the project and any likely damage can be detected and then assessed to determine risks and damage at the population level.

Natural enemies utilized for biological control have co-evolved with their host plant and so are intricately linked to that host plant and do not recognize other plants as potential hosts. Natural enemies have the same mutation rates as native insects, which may also have particular hosts. So, the risks of mutations for natural enemies and native species are the same. However, most mutations are deleterious and are lost from the population. The chance of a natural enemy mutating and being able to attack another plant species is extremely low. It must be remembered that individuals do not mutate. The mutation occurs in the gamete stage (egg and sperm production stage). So individuals cannot simply start attacking other species.

The rationale behind low successes of mutations occurring is because of the complex host-finding and recognition behaviour by insects. For an insect flying around the environment, it must find its host plant among many other plant species. It does this by first smelling the plant (each plant has its own smell based on chemicals in the plant). The insect then flies to the direction of the plant and when it gets closer, it can use visual cues (each plant has a colour spectrum, which is recognized by the insect). The insect then lands on the plant and ‘tastes’ the plant with its tarsi (feet). If the plant seems suitable, the insect may take a bite of the plant. Then if the plant is acceptable, the insect will begin active feeding. However, for the insect to survive on the plant, it must be able to process the plant and breakdown the plant chemicals etc and utilize them in growth and egg development (for females) and sperm development (for males).

So, for an insect (or natural enemy) to be able to attack another plant, it needs to undertake numerous mutations simultaneously. i.e. it needs to be able to recognize a new plant by smell, sight and taste and then also have the enzymes to be able to process the new plant. One mutation on its own will not achieve a change in host.

Even if there is a mutation where the individual can recognize another plant, it won’t necessarily be able to utilize the plant. Also, even if a mutation occurs where the individual can process another plant, it won’t be able to recognize the plant as a potential host, as other mutations have not occurred simultaneously.

Further reading on this topic:

• Futuyama, D. J. (2000). Potential evolution of host range in herbivorous insects. Host-specificity testing of exotic arthropod biological control agents: the biological basis for improvement in safety. R. G. Van Driesche, T. Heard, A. S. McClay and R. Reardon. Morgantown, West Virginia, USA, USDA Forest Service Bulletin: 42-53.

• Van Klinken, R.D. and Edwards, O.R. 2002. Is host‐specificity of weed biological control agents likely to evolve rapidly following establishment? Ecology Letters, 5: 590-96.

The costs of a biological control can vary considerably, depending on how much is known about the weed, its centre of origin (i.e. where it is native) and information about its natural enemies (i.e. whether they have been utilised by other countries already). Costs of projects where information is known can be as little as $10,000 plus costs associated with infrastructure (i.e. laboratories, glasshouses etc may need to be built) and staff. Projects where little is known about the target weed (i.e. it is a novel target, with no previous work conducted), may cost upwards of US$1 million and run for numerous years as the native range of the weed may have to be determined, natural enemies sought, tested, reared and released. Typically, projects aiming to develop new natural enemies may run for at least 10 years.

The cost of one natural enemy is also highly variable, depending on how much information is known about the natural enemy, as well as available infrastructure and staff in a particular country. In the Pacific, natural enemies have been moved around from country to country for as little as US$2,000 if facilities and staff are already in place. In one project, three natural enemies were introduced into one country over three years for a total cost of US$100,000, which included upgrades to facilities and costs for travel to other islands to release the natural enemies.

For other natural enemies, where native range surveys or additional testing may be required, the costs can be considerably higher. Usually, for new natural enemies where no information is known, a cost of US$500,000 is often quoted.

Overall, utilizing existing natural enemies which have been tested and demonstrated to be effective elsewhere, are considerably cheaper than researching new natural enemies.

There are a number of important considerations when rearing natural enemies. The first, and probably the most important, is to have an adequate number of healthy, pest-free plants of the right size and stage of development. Just as we wouldn’t choose to eat poor quality fruit and vegetables, infested with pests, our natural enemies won’t do well on poor quality plants. Mites and aphids may interfere with egg-laying sites. Also, the plants must provide the material or stages that the natural enemy attacks. For example, if the natural enemy feeds on leaves, the plants must have nice healthy leaves, not old hardened leaves. If the natural enemy attacks flowers or fruits, then the plants must have those, otherwise there will be nothing on which the natural enemy can feed.

Another thing to consider is to not keep too many natural enemies in the cage at the one time. They could deplete their food source and then starve and die, or this might result in not enough food for the developing larvae to feed on. For most natural enemies it is a good idea to place adults in a cage with fresh healthy pest-free plants for about one week, before taking them out and placing them into another cage with fresh healthy pest-free plants. This gives time for the adults to lay eggs, but not too much time that they eat all the leaves that the young larvae need to develop. This also keeps generations separate and it is easier to follow the age of the insects, as you don’t have old and new adults mixed together. Old adults lay fewer eggs and are not good for field release.

Finally, it is good to keep accurate records, of when adults are placed in a cage, how many plants were added and when the adults are removed from the cage. By keeping records, it is possible to modify the rearing method if that is needed based on results. It maybe necessary to add more adults or add fewer adults depending on how many eggs are laid or how much feeding occurs.

The most important consideration is to look at the cages daily to ensure the plants and insects are healthy. Wilting plants will affect the performance of the natural enemy, as will pest insects such as aphids, mealybugs etc. Through daily observations, changes can be made before it is too late and you lose the culture.

There are no designated insect rearing training programmes that we are aware. However, staff in countries that import natural enemies to control invasive plants, can receive specific training for that particular natural enemy from experienced researchers who have reared the natural enemy previously. Such training has been provided to researchers in the Cook Islands, Fiji, Papua New Guinea and Vanuatu, who have all introduced natural enemies in the past 10 years.

The benefits of such training is that it is adapted to the particular natural enemy and modified to suit the facilities and conditions in the country. Along with hands-on training, comprehensive notes are also prepared for later reference.

For general information on rearing natural enemies refer to:

• Julien, Mic, and Graham White. 1997. Biological control of weeds: theory and practical application. Australian Centre for International Agricultural Research (ACIAR).

There are numerous textbooks that provide general information on rearing natural enemies. However, for any natural enemies that have been studied previously, it is best to consult the researchers who have had experience in rearing them.

For general information on rearing natural enemies refer to:

• Julien, Mic, and Graham White. 1997. Biological control of weeds: theory and practical application. Australian Centre for International Agricultural Research (ACIAR).

It is best that each rearing cage has only the one natural enemy. There are several reasons for this.

1. In close confinement their actions may interfere with each other. Leaf-feeding insects may destroy oviposition sites for other leaf-feeding insects.

2. Their life cycles may be different and so one may develop quicker than another. This will affect when the plant gets broken down or replaced.

3. The plant parts attacked may be different. For leaf-feeders, there is a requirement to have lots of branches to produce many leaves. Stem borers may want single stems. Flower and seed-feeders need lots of flowers and seed heads. There is a need to grow plants to suit the natural enemy.

4. Some natural enemies may have different temperature and humidity requirements.

5. Having multiple natural enemies in cages makes it harder to monitor their development.

In general, it is not recommended to have more than one natural enemy in a cage. It is possible to have more than one natural enemy in the same nursery or laboratory, as long as they are in separate cages.

Natural enemies will not eradicate the target weed. However, the aim is to suppress weed populations below a level where the weed is no longer causing environmental or economic impact. The number of individuals released to achieve establishment varies with the natural enemy. Some natural enemies, e.g. the beetles on water hyacinth can be established with as few as 50 individuals. Other natural enemies, e.g. the chromolaena moth require thousands to achieve establishment. From experience we have rules of thumb available as to an appropriate number to release for each species or type of natural enemy.

Once the natural enemy has established in the environment, populations will build naturally and hopefully reach levels to control the weed. Generally, it is not feasible to release the numbers required to quickly achieve control, as this could be hundreds of thousands. Natural population growth is needed. So, the aim is to release enough individuals to gain establishment and allow natural population growth to occur to achieve control of the weed.

There are various challenges with implementing any biological control programme. The first challenge is finding a natural enemy that is suitably specific to the target weed and will not attack other plant species. Sometimes this happens and we are unable to release a natural enemy. Another challenge is finding natural enemies that are suitably damaging. Unfortunately, we can’t always predict how natural enemies will perform in their new home. Some natural enemies either do not establish or cause little impact to their target weed.

Once there is a natural enemy ready for release, approval to introduce it into a country needs to be obtained. This could take months, or even years. When releasing natural enemies into the environment, consideration is given to how many individuals to release (varies with natural enemy), time of day (may need to avoid hot midday sun), humidity (need moisture for pathogens), healthy target weeds, sites which are not going to be cleared or are not prone to flooding or other disturbances. There is also a need to ensure that plants in the field are at the right stage and have the particular plant parts needed, e.g. if a seed-feeding insect is going to be released, then there will need to have seeds available etc

The main issue to be aware of when moving natural enemies from one place to another is to keep them cool and out of the sun. Insects and plants should ideally be transported in cooler boxes or similar. Hot cars are just as detrimental to insects and plants as they are to children and small animals. Even once the new release site has been reached, keep the insects cool. Please do not place the containers on the bonnet of a car, on the road or on the ground in the sun.

Ideally, releases should be made in the cooler parts of the day. However, this is not always possible. Keeping the insects healthy will increase the chance of establishment.

The timeframe to monitor and evaluate natural enemies varies with the natural enemy and the target weed. Natural enemies which have a very quick life cycle and rapid rates of population increases need to be monitored sooner than those natural enemies that have a long life cycle and a slow rate of population increase.

The first step of any monitoring programme is to determine whether or not the natural enemy has established. For beetles and moths, it could be whether eggs, and/or larvae can be seen a few weeks after the natural enemy was released. For internal feeders such as gall formers, then looking for signs of symptoms on the plant, such swellings on stems, leaves or flowers, depending on the site of attack. For pathogens, it could be checking for new pustules on the plants in the field.

It is a good idea to follow up this first check with a second look to determine if the insects and pathogens are still present a few weeks later. This is because sometimes there can be a flurry of activity following release but numbers do not persist or larvae do not complete development and a second generation is not produced. Establishment can be confirmed one the natural enemy has been through several generations and appears to be increasing. Checking sites regularly especially if they are easily accessible, means one can see what is happening better than checking sites many months apart. The biggest mistake made is insufficient and infrequent checking of sites, after the natural enemy has been released.

Once there is confidence that the natural enemy has established a viable population, regular monitoring can occur where numbers of natural enemies and impact on the target weed can be assessed. The sort of information collected depends on the natural enemy, the target weed and what sort of information is required.

See response to the question above. The most appropriate time to start monitoring the impact of a natural enemy is once there is confirmation that the natural enemy has established. Depending on the type of natural enemy, this could be a few weeks to over one year. In general, once there is confidence that the natural enemy has established a viable population, regular monitoring can occur where numbers of natural enemies and impact on the target weed can be assessed. The sort of information collected depends on the natural enemy, the target weed and what sort of information is required.

The length of time for a natural enemy to achieve impact can vary, depending on the natural enemy. Some natural enemies can control a weed in less than 2 years. Others may take longer. There have been numerous examples where the water weeds have been controlled in less than 3 years. Chromolaena in Papua New Guinea was controlled in some places in 3-4 years while it took 4-6 years in other places. Climate has a big influence of how long control may take.

It is advisable to regularly monitor a release over the months following a release. First this can determine first if the natural enemy has established and second if natural enemy numbers are increasing and plant health is decreasing. Normally, it is advisable to monitor for at least a full 12 months, which covers the four seasons and any seasonal differences in climate.

In theory, it may be possible to use time-lapse cameras to capture control of a weed over time but it is likely to be difficult to do because of the time frames required Some natural enemies have been able to control a weed at certain sites in less than 2 years, but this is too long to be feasible for time-lapse photography

No. Time-lapse cameras have not been utilized to date. Up to now, photos from a set point have been taken over a period of time. This works similarly to time-lapse cameras. The frequency of photos may depend on the natural enemy as well as access to the site. For some release sites, photos were taken every three months, while for more distant sites, photos may have only been taken once a year. A series of photos from a set point will be placed on the NENS webpage, as an example.

The best way to determine if a natural enemy is working is through observations captured by camera as well as scientific monitoring through data collection. The latter involves using making measurements of plant traits such as abundance, density, and height against the abundance of the natural enemy. This is conducted over time to show trends indicating that the natural enemy is reducing the abundance, density or height of a target weed.

Another measure to show a natural enemy is working is through socio-economic surveys. These surveys are important as it captures landholder perceptions of the target weed. It is possible for instance that monitoring through data collection and photos might show that a weed is becoming less abundant. However, a landholder may still believe that the weed is having a negative impact on their crops or farmland. This suggests that adequate control of the weed has not been reached. Using a combination of all three techniques (photos, data collection and surveys) enables a better understanding of the system and whether or not additional natural enemies need to be released to achieve the desired level of control.

Our research focuses on invasive alien plants which are normally controlled using insects or plant pathogens although mites are occasionally used as well. Therefore, we do not undertake research on crown of thorn starfish or other animals. Such research is undertaken by other research groups.

The most critical attributes to consider when choosing a natural enemy is that it must be specific to the target weed and not be a risk to other plant species. Detailed host specificity testing will go a long way to address this, as well as knowledge of the natural enemy in other countries, especially if it has been established in other countries for many years.

A second factor is that it must be capable of damaging the target weed. For many natural enemies introduced into the Pacific, they have been released in other countries and so not only has their specificity been determined but so has their effectiveness. For instance, over 40 natural enemies have been released against lantana. However, while the natural enemies do not attack other plant species, through field observations in other countries, only about 6 or 7 of these natural enemies would be recommended for release in other countries because of their ability to contribute too lantana control.

There are several traits considered desirable for effective natural enemies. Long-lived adults, where females can produce lots of eggs over a long period of time is useful as populations can build up quickly. A short life cycle i.e. the time it takes the natural enemy to develop from egg to adult is desirable as more populations can occur over a year, also allowing for a rapid build-up of numbers. Another trait is that the natural enemy must cause significant damage to the plant such that the plant becomes weakened or produces fewer flowers or seeds. Some natural enemies may feed on their target weed but the damage is not significant and the plant can still produce large quantities of flowers and seeds. It is also helpful, but not essential for natural enemies to disperse readily and rapidly. Effective natural enemies are those which can escape high levels of parasitism and predation, and thrive under the climatic conditions available.

In general, as long as the natural enemy is being released in similar environmental conditions as where it has proven to be effective, it should be effective in a new country. Natural enemies released in one country in the Pacific should do equally as well if released in another country in the Pacific. However, it is important to remember there can be subtle differences, even within countries. An example is the gall fly on chromolaena is effective in areas which are not too dry or not too wet. In the drier regions of PNG, the gall fly took a little longer to control chromolaena than it did in the coastal regions. Control was also slower in West New Britain where it is very wet and cloudy. As the gall fly prefers sunny warm areas, the conditions in West New Britain were not as conducive for the insect to thrive and build up into damaging populations, as in areas in which the gall fly was able to control chromolaena. Another example is with the mikania rust. The rust requires high humidity and cool nights (<25>

There are standard protocols when introducing new natural enemies into a country. FAO and other organizations have published these to provide a minimum standard. These protocols include collecting natural enemies from other countries, packing, transporting and introducing into a country.

Further reading on this topic:

In most countries importing natural enemies for biological control is treated the same as importing any commodity item and goes through the same risk assessments. Some countries such as Australia and New Zealand have very well-developed systems and have experienced people to be able to make such assessments. Other countries such as Papua New Guinea and Vanuatu which have a long history of introducing natural enemies, also have good risk assessment procedures but may rely others to assist with the assessments. For instance, both of these countries base their risk assessment document (and the information that it contains) on the ones used in Australia.

Other countries which have never introduced a natural enemy before, may not have a system in place to do so. In the Pacific, 17 of the 22 Pacific Island countries and territories (PICTs) have deliberately introduced a natural enemy against a weed species. In addition, every one of the 22 PICTs have deliberately introduced a natural enemy against an insect pest. So, there is a system in place for each country to introduce natural enemies but as some countries have not introduced a natural enemy for over 25 years, the people required to process such applications now may not have the necessary experience or knowledge, and may require some training or support Independent expert advice may also be helpful and this is available if needed.

Typically, if a country wants to introduce a natural enemy, an import risk assessment is prepared. This document should contain information on the target weed, including its impacts, as well as any possible benefits, and information about the natural enemy, including its biology, host specificity and its usefulness as a natural enemy in other countries. Examples of a typical risk assessment can be placed on the NENS website.

Once the document has been assessed, import permits can be issued, to allow the natural enemy to be brought into the country.

In short, yes. This is because much of the information relating to the testing and effectiveness of the natural enemy will not vary from one country to another. However, it is worth revising the section on the target weed for a particular country. Under this section, closely related plants to the target weed found in the country are listed. It may be possible that there could be a plant species closely related to the target weed found in one country but not another and has not been tested against the natural enemy. In this case, additional testing on that plant species may be necessary prior to introducing the natural enemy.

IRAs prepared for the introduction of natural enemies in Fiji, Papua New Guinea and Vanuatu were all based on IRAs used in Australia. A template has now been prepared for use for all Pacific countries and can be modified for a particular country. Countries wishing to import a natural enemy that has already been introduced into another country, can use the existing IRA and modify it accordingly. This will save a lot of work, much of which would be repeated.

A template has now been prepared for use for all Pacific countries and can be modified for a particular country. This is now on the NENS Section of the SPREP website. Existing IRAs have also been placed on the website as a guide. Information of the target weed usually is obtained from the particular country through herbaria or local knowledge. Information on the natural enemy can be obtained from the scientific literature via Google Scholar.

Information can also be obtained by emailing researchers with knowledge of the target weed and the natural enemy.

It is worth noting that for the vast majority of natural enemies, IRAs would have been prepared for other countries and general information can be extracted from those documents.

The rationale behind introducing natural enemies from other countries is because the weed was introduced from another country, without its natural enemies. Insects and pathogens in the country of introduction are not adapted to attack the weed. So, without its natural enemies, the weed becomes problematic as there is nothing to keep it under control. Hence, it is necessary to bring in natural enemies from the weed’s country of origin to restore the balance.

These natural enemies collected from the target weed are first thoroughly tested to ensure that they will not pose a risk to other plant species in the country of introduction. For many weeds in the Pacific, natural enemies introduced to control a weed species have not only been thoroughly tested and approved acceptable for introduction but have already been tried and proven in other countries. So, there is long-term data supporting evidence that the natural enemy is not going to pose a risk to other plant species.

The system worked well for the Cook Islands. A similar system has been adopted by stakeholders in New Zealand and South Africa.

Yes, they can be. These usually arise when a weed may be a problem to some sectors of the community but valued by others. A classic case is with the fast-growing vine, mile-a-minute, Mikania micrantha. The plant can smother other vegetation, reducing flowering, fruiting and yields of papaya, bananas, cocoa, coffee, coconut trees and oil palm. It can smother taro and green, leafy vegetables. However, the plant is also used for treating cuts and other ailments. Some farmers say that the weed is good to keep out other weeds and is useful for mulch.

Fortunately, natural enemies don’t eradicate plants and so they will always be some plants available for any beneficial uses.

Most countries have separate departments managing the environmental sector and the agricultural sector, with quarantine, often falling under the Department of Agriculture. As invasive plants often impact on agriculture and the environment, it is wise to involve both departments in any approval process, both for targeting weeds for biological control or releasing natural enemies. It serves two purposes. First, it ensures there is no conflict of interest. For instance, a plant that may become troublesome in food gardens, could actually be a native species. The reverse could also apply, where a plant which is invading the environment, could have some agricultural uses.

Similarly, when gaining approval to release a natural enemy, both jurisdictions should be involved to ensure that the natural enemy isn’t going pose a problem to agricultural plants or to the environment.

Not that we are aware of. In most cases, the community will rightly raise concerns and ask about the risks in establishing a NENS programme. Researchers and the relevant department officers will then attempt to address their concerns. History shows that 17 of the PICTs have deliberately introduced a natural enemy. So, there has not been a blanket rejection of a NENS programme. However, as mentioned above, there may be some reluctance to engage natural enemies to target particular species which are deemed to have benefits, e.g. Mikania micrantha. This is where it is important to have public consultation to ensure all concerns are addressed.

There was a NENS programme in Timor Leste in the early 2000s. As Timor Leste had never undertaken a NENS programme before, there was some initial reluctance by some sectors of the community. As with other countries who have had little experience with NENS programmes, there was concern of non-target impacts by natural enemies on other plant species. The most important action to take is to engage the community and listen to their concerns. Open dialogue and information sharing is important at an early stage.

The next step is to address each one, with the aim to reassure and provide evidence that such non-target impacts are highly unlikely to occur as the natural enemies have not only been thoroughly tested but have been released in numerous countries previously, all showing that there were no non-target impacts on other plant species.

Through community engagement, two natural enemies that had been utilized in other countries were released in Timor Leste and there have since been discussions on introducing other natural enemies to target other weeds. There is one publication related to this project.

• Day, M.D., Brito, A.A., da Costa Guterres, A., da Costa Alves, A.P., Paul, T. and Wilson, C.G. 2013. Biocontrol of Chromolaena odorata in Timor Leste. In: Proceedings of the Eighth International Workshop on Biological Control and Management of Chromolaena odorata and other Eupatorieae. (Eds. Zachariades, C., Strathie, L.W., Day, M.D. & Muniappan, R.), pp. 134-140. ARC-PPRI, Pretoria.

Not that we are aware. In most cases, any conflicts of interest have been addressed without a cost-benefit analysis and the NENS programme was able to proceed. In the one case that we are aware of, we made a decision to not pursue a NENS programme in a particular country until the conflicts of interest had been resolved. Here, it is up to the particular country to resolve this.

Not necessarily. For many weeds, the impacts are already well known and there is general agreement to introduce a host specific and effective natural enemy that has already been utilized in another country. In Vanuatu, there was widespread agreement that Mikania micrantha is one of the most important weeds in the country and there was overwhelming support to introduce the rust which had been released in numerous other countries including PNG and Fiji.

However, cost-benefit analyses may be undertaken to clarify or address conflicts of interest or assist in determining which particular weed is the most important weed to target. This more often occurs in countries such as Australia, New Zealand and South Africa where new NENS programmes are being developed and initial costs of the project can be high.

However, programmes which are funded by external agencies, often have a requirement at least to document the impacts of a target weed as baseline data, so the economic, social and environmental benefits of that programme can be captured in years to come.

Weighing up the pros and cons of introducing a natural enemy is usually conducted during the consultation phase of the programme. Here the status of the weed can be assessed to determine its impacts on agriculture and the environment and if there are any benefits of the weed. It is also important to consider the cost of conventional control or if the plant can be eradicated.

If the target weed is causing high impacts to agriculture or the environment, and there is one or more highly specific natural enemies that are causing damage to the same species in other countries, then it may be very cost-effective to introduce a natural enemy to control a weed, rather than continue with conventional control methods.

Conversely, if the weed is not causing high impacts to agriculture or the environment, then pursuing biological control may not be the best option. Similarly, if the natural enemy attacks other species, or there needs to be a great deal of extra testing, introducing a natural enemy may not be cost-effective. Finally, if the natural enemy does not have a good track record of causing high impacts to the target weed, then introducing it may not be beneficial.

Once a weed has been agreed to be the target of biological control, the pros and cons of introducing specific natural enemies can be discussed. For some weeds, there may only be one natural enemy available, but for other weeds such as lantana, there could be up to 20 natural enemies.

In both situations, the specificity of the natural enemy needs to be taken into consideration i.e. does it attack other species, even if only slightly, and are there additional plants that need to be tested prior to its introduction into a specific country. The impact of the natural enemy or enemies also need to be assessed. Are the natural enemies likely to control the target weed? If there is more than one natural enemy available, is one more damaging than another? All these issues need to be considered.

Certainly. For some countries, the most important weed is already known. This could be from previous workshops or other means. When the mikania rust was introduced into Vanuatu 10 or so years ago, it was because the various organizations, including Agriculture and the Environment in Vanuatu deemed Mikania micrantha the most important weed. For the case with mikania, there was no disagreement or conflict of interest.

The consultation process is a way to identify any conflicts of interest or to determine if other weed species may be of higher importance. For some countries, who have not undertaken biological control ever, or for many years, the consultation process helps prioritise.

Certainly. For some countries, the most important weed is already known. This could be from previous workshops or other means. When the mikania rust was introduced into Vanuatu 10 or so years ago, it was because the various organizations, including Agriculture and the Environment in Vanuatu deemed Mikania micrantha the most important weed and they supported the introduction of the rust. For the case with mikania, there was no disagreement on its importance or conflict of interest.

The consultation process is a way to identify any conflicts of interest or to determine if other weed species may be of higher importance. This is partly because different departments have different priorities. For instance, the most important weed for the Department of Agriculture may not be the most important weed for the Department of Environment. For some countries, who have not undertaken biological control ever, or for many years, the consultation process helps prioritise.

Yes. Many countries know what their most important weed species are. In these cases, it is not necessary to have a list of weed species from which to prioritise. The consultation process however, can determine if other stakeholders also believe that this target weed is a priority. For instance, one department may determine a weed species is a priority but another department may determine the status of the species is not suitable for biological control. For instance, a plant may be deemed to be native or a plant may be deemed to be useful to a select group of farmers or interest groups.

An example was when a particular country was considering introducing the mikania rust. There was general consensus that the rust should be introduced. However, during the consultation process, there was a number of stakeholders that said that mikania is useful as a cover crop to reduce other weeds and for the treatment of cuts. The group decided that it was not a good idea to pursue introducing the rust until there was wider community support.

The scale of the consultation is usually decided by the relevant organizations involved in the project. The in-country people are going to have a better idea about what stakeholders exist in a country or island. It is not up to the researchers, although they can provide advice on who could be included, and can assist if required. For some species which have impacts on limited areas or environments, there could be less consultation than weeds that may impact across the board. For instance, there may be less consultation with aquatic weeds than weeds that affect agriculture, natural areas or may be used in herbal medicine. Usually, the consultation process draws out the impacts and possible uses of a particular weed. Hence, why there should be broad scale engagement with the community. For instance, there could be a small group of the community that use a weed. The consultation process needs to be wide enough to capture these stakeholders.

Occasionally. During the consultation process, various stakeholders share their views regarding the value of a weed species and whether the introduction of natural enemies should be undertaken. All views must be respected. Open and constructive dialogue is essential as some views might be based on false logic or misleading information. In the end if there is opposition to the introduction of a natural enemy which cannot be resolved, then the programme does not proceed. This was the case with the mikania rust in one country. As it didn’t get widespread community spread, the rust was not introduced.

Occasionally. There was opposition by some stakeholders to the introduction of the rust to control Mikania micrantha. Due to this opposition, the project did not proceed.

During the consultation process, various stakeholders share their views regarding the value of a weed species and whether the introduction of natural enemies should be undertaken. All views must be respected. Open and constructive dialogue is essential as some views might be based on false logic or misleading information. In the end, if there is opposition to the introduction of a natural enemy which cannot be resolved, then the programme does not proceed.

It only took a few months. This was helped by a range of already published reports on weeds in the Cook Islands and the Cook Islands Biodiversity & Natural Heritage Website database (http://cookislands.bishopmuseum.org/search.asp), which identifies the most seriously invasive weeds (179 of 333 exotic plants present in the Cook Islands were listed as “invasive” of which 46 were described as “serious”). A shortlist of the 46 “serious” invasive plants was scored in terms of suitability for biocontrol (based on predicted cost and impact of a biocontrol program). A workshop was subsequently held in Rarotonga where 12 local delegates with interests in forestry, horticulture, livestock, biodiversity conservation & biosecurity scored weeds by importance and an overall prioritisation combined the weed importance and biocontrol suitability scores (where the best targets were weeds that were both important and considered to be good biocontrol targets).

Yes. Many countries in the Pacific have similar landscapes and industries. So, using lists prepared by other countries or organisations makes sense. However, it is best to carefully check and modify for your own country or region. There could be small idiosyncrasies that may change the priority of one weed over another. For instance, there may be a major industry in one country which is not so major in another country. If a weed is important to the first country as it impacts on that major industry or commodity, it may be less important in the other country. For example, Sida spp. were major pasture weeds and obvious biocontrol targets in Vanuatu but in the Cook Islands, which does not have a major cattle industry, Sida is not considered to be a major weed.

Also, a weed may be important in one country and not have any natural enemies while the weed in a second country already has those natural enemies. This will also change the priorities between countries.

So, while it is useful to use other priority lists, it is important to understand the context and reasons why the weeds were ranked as such in another country.

Yes. This document is already on the NENS section of the SPREP website. The prioritization exercise has been completed for 90 weed species. As there are over 1600 weed species present in the Pacific, many species have not been assessed yet. Fortunately, many of these species, are not invasive or are not deemed a high priority. We are looking at calculating scores for other weed species that are deemed a priority in various countries. Countries that wish particular species to be assessed, could contact Quentin Paynter at PaynterQ@landcareresearch.co.nz

The primary role of a country supplying a natural enemy is to supply the latest up-to-date information on that natural enemy. This includes information on its biology, specificity and also its impact to the target weed. The information should be as detailed as possible to help the country conducting the assessment make a sound decision as to whether or not to introduce a particular natural enemy. The two most critical aspects are, does the natural enemy attack or feed on any plant species other than the target weed and is the natural enemy likely to provide substantial control to the target weed.

Other aspects could include information on how to rear, field release and monitor the natural enemy. Countries wishing to introduce a natural enemy may wish to request scientific publications to substantiate and support information supplied.

Determining whether a particular weed has any value in the local community can be achieved through a consultation process. By engaging a wide range of stakeholders, it is possible to determine if there are any groups in the community that may have uses for a weed. For many weeds, e.g. Mikania micrantha, uses are often known. However, it is possible that for some weeds, a small section of the community may have a specific use of a weed that is not known to the broader community.

In general, land tenure issues are not a significant problem. However, it may happen where landholders might be concerned that the natural enemy will attack their crops and they do not want the natural enemy released on their land. Most land managers are happy to have a safer, more sustainable solution for managing weeds, once they understand what is involved.

In some countries, managers of some national parks have expressed concern regarding the release of natural enemies in protected areas. However, natural enemies, once established, will move and cross boundaries (property, local government or state boundaries). So any issues of this nature should be addressed during the consultation process prior to the release of the natural enemy in the country.

The length of time to complete a NENS depends on numerous factors. The time can be relatively short (a few years) if the natural enemy has already been tested and is known to be highly effective. Several natural enemies which had been tested and utilized elsewhere and then released in Vanuatu, achieved control of their respective weeds within 3 years.

However, for novel weed targets where there are no known natural enemies, the time frame can be considerably longer. The biological control programme for parthenium weed took over 20 years before control was achieved and adequate control of lantana is still being sought for some situations despite over 100 years of research. The longer times are because surveys need to be conducted in the native range of the weed to locate potential natural enemies, before host specificity testing can be conducted to demonstrate that they are suitable for release. Even then, there are no guarantees that the natural enemies released will actually control the weed. For example, many of the natural enemies on lantana cause only slight damage to the weed.

There are no barriers as such. The important factor is to ensure that the consultation process is thorough enough to involve all those with vested interests. Most of the problems arise when all available information hasn’t been made public. It is important to know if particular weeds have a use for one group, no matter how small. It is also important to have representatives from environment, agriculture and forestry so that the group know all the weeds that are causing problems and can work through all the issues to enable a priority list to be prepared based on all information. If some sectors are not represented, then it is possible that important information is overlooked.

The stigma on the use of natural enemies is not limited to just the Pacific region. Some countries in Africa and Asia have also expressed reluctance to implement biological control. This is despite some of these countries having introduced natural enemies before, resulting in substantial control of the weed, without any negative side-effects on other plant species. Some of these countries also quite willingly will introduce natural enemies against insect pests of agriculture.

There are several possible explanations. The first is that countries may not have ever introduced a natural enemy before and are unaware of the theory of biological control and use of natural enemies. The countries may also not have the processes in place or facilities and expertise to be able to introduce a natural enemy.

In other situations, countries or individuals within countries are highly risk-adverse and may think the natural enemy will attack other plant species such as crops through mutations or evolution or when the weed is controlled. There is often a lack of knowledge on plant-insect interactions and how natural enemies have co-evolved with the weed. See sections under host specificity for further information.

Another explanation is that it is commonly known that past ill-advised introductions of generalist species such as myna birds, mongooses, giant African snails etc have caused much harm. However, this should not be confused with the carefully considered of specialist natural enemies for weeds, which has been done safely and successfully for more than century, and has delivered many benefits to the region.

For further reading:

• Day, M.D. & Witt, A.B.R. 2019. Weed biological control: Challenges and opportunities. Weeds – Journal of Asian-Pacific Weed Science Society 1, 34-44.

As the natural enemies are specific to a particular weed, if there are two weeds, different natural enemies for each weed will need to be released. If only natural enemies for one weed are released, then the other weed might increase in numbers once the targeted weed is controlled. In the Cook Islands, three invasive vines were targeted simultaneously because controlling one vine will result in the other two vines invading the spaces occupied by the weed that was controlled.

This practice has been used in other countries where controlling multiple weeds invading a particular habit such as pastures or aquatic systems is seen as highly beneficial. In PNG, three aquatic weeds all had natural enemies released so one does not take over as another is controlled.

There is also a current project where three weeds of pastures are being targeted simultaneously in Vanuatu for the same reason.

While natural enemies are better deployed for those weed infestations that are too large to manage by other means, there is scope to utilize natural enemies to help contain weed populations when they are localized and haven’t become too problematic. However, before introducing a natural enemy which will involve risk assessments and importing the natural enemy into a quarantine facility, consultation is recommended to determine the options available and to weigh up the pros and cons of introducing a natural enemy versus undertaking other management strategies.

A good example of where natural enemies can be deployed against weeds which are localised is with some aquatic weeds which may only be found at one or two sites which are too large to eradicate using chemicals or by manual means. Also if the weed did spread to other sites, then the natural enemy is already in the country and can quickly be utilized to prevent further spread.

Some sap-sucking insects can transport disease. There has been anecdotal evidence that the feeding actions of the lantana bug causes death to parts of the lantana plant away from the feeding site.

However, it is important to recognize that the insect is only transmitting the disease from one lantana plant another, as the insect feeds only on lantana.

At present there is no programme for natural enemies for little fire ants. Currently populations are being treated through a baiting programme.

For giant African snails, natural enemies have been investigated but as far as we are aware, no one has been able to find a natural enemy that is sufficiently specific to attack only giant African snails.

In general, the ability of natural enemies to adapt to climate is a good thing. If climate is changing and the weed distribution is increasing as a result, then it is beneficial for the natural enemy to be able to adapt to these new environments as well. Otherwise, the weed may become out of control. It is unlikely that a natural will enemy will evolve or become adapted to areas unsuitable as the weed needs to be there. Therefore, the natural enemy won’t be there either.

In Australia, as it warms, the distribution of lantana spreads further south, as cooler areas warm. However, if the natural enemies can also adapt, then they can help control lantana in these new areas. Similarly, if the weed is already found in a warm climate and it becomes hotter and the weeds survive under the new conditions. If the natural enemies can also survive the hotter conditions, then the weed may not become a problem.

Yes. Eighteen natural enemies have been deliberately released against nine weed species. Of these natural enemies, 11 have established on eight weed species. An additional three natural enemies have established in the Cook Islands, having spread from other countries, possibly on plants or by wind. Of the natural enemies that have established in the Cook Islands, six are causing variable to high impacts on their respective target weed (including good control of Lantana, grand balloon vine and giant sensitive plant and variable levels of control of mile-a-minute vine). Several natural enemies have only been recently released and their impact on their target weeds is yet to be quantified. Details can be found in Day and Winston 2016 or on-line at https://www.ibiocontrol.org/catalog/

Yes. The NENS section of the SPREP website and @PacificNENS Facebook site have been set up for this purpose. On the website you will find a paper reporting which natural enemies are present against each weed in which of the PICTs and also how effective each natural enemy is. Also a searchable database which provides greater information on what natural enemies are present in each of the PICTs to help countries select potential natural enemies for particular weeds.

We ask countries if they have new information on a particular natural enemy to share it with us so the databases are updated for others considering introducing natural enemies for a particular weed.

• Briese, D.T. 2005. Translating host-specificity test results into the real world: The need to harmonize the yin and yang of current testing procedures. Biological Control, 35: 208-14.
• De Clercq, P., Mason, P.G. and Babendreier, D. 2011. Benefits and risks of exotic biological control agents. BioControl, 56: 681-98.
• Futuyama, D. J. (2000). Potential evolution of host range in herbivorous insects. Host-specificity testing of exotic arthropod biological control agents: the biological basis for improvement in safety. R. G. Van Driesche, T. Heard, A. S. McClay and R. Reardon. Morgantown, West Virginia, USA, USDA Forest Service Bulletin: 42-53.
• Hinz, H.L., Winston, R.L. and Schwarzländer, M. 2019. How safe is weed biological control? A global review of direct nontarget attack. The Quarterly Review of Biology, 94: 1-27.
• Paynter, Q., Fowler, S.V., Gourlay, A.H., Peterson, P.G., Smith, L.A. and Winks, C.J. 2015. Relative performance on test and target plants in laboratory tests predicts the risk of non-target attack in the field for arthropod weed biocontrol agents. Biological Control, 80: 133-42.
• Suckling, D.M. and Sforza, R.F.H. 2014. What magnitude are observed non-target impacts from weed biocontrol? PloS one, 9: e84847.
• Van Klinken, R.D. and Edwards, O.R. 2002. Is host‐specificity of weed biological control agents likely to evolve rapidly following establishment? Ecology Letters, 5: 590-96.
• Winston, R.L., Schwarzländer, M., Hinz, H.L., Day, M.D., Cock, M.J.W. and Julien, M.H. 2014. Biological control of weeds: a world catalogue of agents and their target weeds. USDA Forest Service, Forest Health Technology Enterprise Team, FHTET-2014-04, Morgantown, West Virginia.

There are four organizations in the Pacific region that specialize in NENS work. Two of these are in Australia, the Queensland Department of Agriculture and Fisheries, and CSIRO. However, CSIRO no longer have projects in the Pacific and have not worked in the Pacific for 25 years. Manaaki Whenua - Landcare Research in New Zealand and Department of Agriculture in Hawai’i also currently conduct NENS work in the Pacific.