Releasing biological control agents
In Section 7 of the application (release) or Section 9 (release with controls) the applicant could provide details of the proposed release programme, including information on the breeding and culture (where applicable), and the life-stage and number of the organisms to be released; timing and location(s) of release or release with controls etc. The costs of the importation, rearing and release phases of the programme, as well as handling and harvesting methods involved, could also be covered.
There are many other issues that should be taken into account when designing a release programme, and an applicant may wish to include these in completing the application form. Here is a fuller treatment of the factors that can influence the design of a release strategy, illustrated where possible with New Zealand examples.
Establishing biological control agents is not a straightforward matter. Establishment may be instantaneous, may occur only after a long period, or may not happen at all. History shows us that ease of rearing in the laboratory is no predictor of establishment success.
- Selecting release sites
- Timing of releases
- Release strategies
- Collection and redistribution
- Other issues to consider
Selecting release sites
Agents are usually sourced from areas of the native range that broadly match the area in which the agent is to be established, and sometimes more than one biotype is introduced to manage pests in different climates (van den Bosch et al. 1982). Climate can influence establishment success in a variety of ways. At its simplest, weather can influence biological control agent survival and dispersal at the time of release. Once released, the chosen release site should provide adequate conditions for the control agent to develop through all stages, while temperature, humidity and rainfall extremes should not limit survival of any stage. Inappropriate weather conditions can also interfere with insect behaviours such as mating and oviposition. Where it is uncertain exactly what conditions the agent needs to survive or breed are uncertain, releases should be made over a range of climates to maximise the likelihood that an agent population will survive. Climate can also influence your choice of the best time of year to release agents.
If a pest insect attacks a range of plants, it is advisable to release control agents onto a range of alternative host plant species. Release amongst diverse vegetation rather than a monoculture may assist the agent to use natural host-seeking behaviours that might increase the chance of establishment. Non-crop vegetation also provides a refuge for the agent that is not directly affected by agricultural practices such as pesticide application. Release in permanent vegetation also increases security that the new population won’t be destroyed by harvest, cultivation or grazing.
Habitats can be enhanced in anticipation of the release of control agents. Fertilisers have been applied to weeds to enhance the reproductive potential of the agents applied. Establishment success can be improved by planting additional plants infested with the host into the release site, or simply as an 'out of season' host plant for natural populations of hosts.
Parasitoid host-finding behaviour is often mediated by kairomones generated by damaged plants. Creating infestations within release sites could arrest emigration from the release site once agents are released.
Adult parasitoids sometimes need to feed on nectar as adults to maximise egg-production (see references in Araj et al. 2006). Planting appropriate flower species within the release site could increase the parasitism rate and fitness of the control agents following release, and increase the likelihood of agent establishment.
Timing of releases
A new control agent can only establish if the target life stage of the host is present when the agent is released. Accurate timing of releases is therefore critical to establishment success. Even though univoltine or bivoltine pest species may be present year round, the life stages that are susceptible to the control agent may be present for a relatively short period. Releases must coincide with this window. In weed control, the susceptible plant structure at the right stage of development must be present when the agent is released. For example, Hill et al. (2000) found that the eurytomid seed-feeding wasp only laid eggs into the seeds of Acacia species for 3 weeks in late spring. If adult wasps were released outside this period, no suitable seeds would be available for oviposition. Similarly, Dymock (1987) found that adult ragwort seed fly, Botanophila jacobeae began emerging in late October and early November, 4-6 weeks before ragwort began to flower. During this time they fed in order to mature eggs. Release of new adults in early December, when flowers suitable for oviposition first appear, might not allow sufficient time for reproductive development before the resource became too mature for larval development.
Synchrony can be manipulated by seeding release sites with target pests of the correct stage, by introducing plants to the release site that are susceptible to the target pest out of season, or by judicious pruning to alter host plant phenology.
Seasonality and synchronicity are closely linked concepts. The period of the year during which agents can be released may be limited by the need to expose the control agents to climatic requirements that condition insects for successful diapause or hibernation. For weed control agents, it may be necessary to release agents during a narrow window when host-plant quality is adequate for successful development.
Hill (1982) found that the protein content of gorse foliage was generally low, peaking for only a short period in the spring. The life history of most foliage herbivores was keyed to this resource. Strategies for releasing foliage-feeding biological control agents in New Zealand were designed with this in mind (e.g. Hill et al. 1995).
Where an exotic agent is already established, or where native species are active, newly-introduced agents might face competition for hosts. Where possible, release multivoltine agents at a time of year when competition is at a minimum. Similarly, predators and host disease can interfere by removing susceptible stages of the pest, or even destroying parasitised individuals. Where such interactions are known, time releases to minimise such interactions.
The release strategy depends on the biology of the agent and the host, and on how many agents are available for release. Characteristics such as high reproductive rate and dispersal will favour establishment from a low number of founding individuals, whereas more sedentary agents may need to be released in larger numbers. It will vary from agent to agent, and no two strategies will be alike. The following principles should be taken into account when planning the strategy.
How many release sites?
Every release site will vary in such attributes as vegetation, aspect and microclimate. Releasing at only one site risks failure not because agents cannot establish in New Zealand but because the particular site was unsuitable, or was destroyed by some stochastic event such as a rainstorm or fire. The more sites that are selected, the more likely it is that at least one will match the agent's requirements for establishment. Memmott et al. (1998) found that gorse thrips (Sericothrips staphylinus Haliday) established at only 33% of the sites where 10, 30 and 90 thrips were released initially, but at 100% of sites where releases numbered 270 or 810 thrips. If faced with limited numbers of agents to release, the first priority is to release sufficient agents to avoid Allee effects (where the reproduction and survival per individual is lower in small populations than in large populations) but otherwise maximise the number of sites at which the agent is released.
How many releases and how often?
There are several release strategies that can be employed. The simplest is to release as many agents as practicable at a release site at the correct time of the year. Alternatively, smaller releases may be made at regular intervals to make sure that agents are in synchrony with hosts, or to augment populations at intervals to overcome Allee and founder effects (Etzel and Legner 1999).
How many agents per release?
In Canada, Beirne (1975) found that on average, the more parasitoids that were released, the more likely it was that the agents established. For species where there were more than 800 individuals per release, 60% of species established, whereas for species where there were fewer than 800 individuals per release, only 15% established. In New Zealand, Cameron et al. (1993) found that where over 30,000 individuals were released over the course of a biological control programme, 80% of species established, but for programmes in which fewer than 1,000 individuals were released, only 44% of species established. The message from both studies seems to be that the more agents that can be released the better. However, release of large numbers is not a prerequisite for agent establishment. Cameron et al. (1993) record that Lathrolestes luteolator, an ichneumonid parasitoid of Caliroa cerasi established successfully in New Zealand even though only 27 individuals were released. The parthenogenetic braconid Pholetesor pedias Nixon was sent to Canada from New Zealand for the control of Phyllonorycter blancardella (F.), but only two females survived the journey (Richard Hill pers. comm.). These were released onto sleeved branches, and within 18 months years, 25.7% of the leafminer larvae in the orchard were parasitised and the agent had been detected 43km away (Laing and Heraty 1981).
What stage to release?
You may have a choice of releasing adults or immature stages. Your choice will depend on the biology of the agent. You may wish to ensure that adults exhibit natural host-searching, mate selection and dispersal behaviours by having them emerge in the field rather than the laboratory. Conversely, you may have few agents, and you may wish to ensure that adults are fertile by having them mate or feed before release. You may wish to hold adults until any pre-oviposition period is over to ensure that most adults lay eggs before dying in the field.
How to release?
Your chosen release method may be as prosaic as driving to a single release site and opening a box. It might be as spectacular as the cassava mealy bug programme in Africa, where agents were released from planes on a landscape scale.
You may choose open release or you may make your first release onto hosts in sleeved stems, or onto plants in cages. There are many cage designs, including "open" cages (delayed emigration). It is important to retain the released agents near the release point to foster local population development. For highly active and vagile agents it may be important to minimise the dispersal instinct of agents. If possible, do not release insects in strong direct sunlight as you risk agents dispersing towards the sun on release. Choose a cloudy day, and release early in the morning or late in the afternoon. Agents with a high propensity to disperse may require larger releases than more sedentary agents.
Some other logistic issues to think about are:
- Boxes: ventilation is often not essential. It is important to keep relative humidity high, but it is also vital to avoid condensation. Incorporate lots of absorbent paper in packaging to absorb free water.
- You may need to feed agents in transit with a honey and water solution delivered on cotton wicks, or sugar solution in solid agar drops.
- Keep insects cool, but not cold.
- Keep all insects out of direct sunlight. Even short exposure is fatal to boxed insects.
- Weather: do not release when temperatures are too low to allow insects to seek shelter, or too high to allow normal behaviours. Do not release in the rain.
Collection and redistribution
The most cost effective way to establish new populations is to harvest from sites where the agent is already present and populations are expanding, and transferring to a new site. It is very unlikely that populations will have increased sufficiently for this to be possible within 2 years of first release, and it may take much longer before this method can be employed.
Other issues to consider
- It doesn't matter how many agents you release, or at how many perfect sites you release at if the agents are of poor quality.
- It may be possible to tailor the selection of individual biotypes for varying habitats/climates
- Rearing, resting stages, mass-production, disease
- Refer to standard texts for more information on these other issues that may be relevant to completing the relevant section of the application forms.
Araj S.A., Wratten S.D., Lister A.J. and Buckley H.L. (2006). Floral nectar affects longevity of the aphid parasitoid Aphidius ervi and its hyperparasitoid Dendrocerus aphidum. New Zealand Plant Protection 59: 178-183.
Beirne B.P. (1975). Biological control attempts by introductions against pest insects in the field in Canada. The Canadian Entomologist 107: 225-236.
Bellows T.S. and Fisher T.W. (1999). Handbook of Biological Control: Principles and Applications of Biological Control. Academic Press, San Diego.
Cameron P., Hill R.L., Bain J., Thomas W.P. (1993). Analysis of importations for biological control of insect pests and weeds in New Zealand. Biocontrol Science and Technology 3: 387-404.
Dymock J.J. (1987). Population changes of the seedfly, Pegohylemyia jacobaeae (Diptera: Anthomyiidae) introduced for biological control of ragwort. New Zealand Journal of Zoology 14:337-342.
Etzel L.K. and Legner E.F. (1999). Culture and colonisation. Pp. 125-197 In: Handbook of Biological Control: Principles and Applications of Biological Control, T.S. Bellows and T.W. Fisher (Ed.) Academic Press.
Hill R.L. (1982). Seasonal patterns of phytophage activity on gorse (Ulex europaeus) and host plant quality. Pp. 237-242. In: Proceedings of the 5th International Symposium on Insect-Plant Relationships, J.H. Visser and A.K. Minks (Ed.) Wageningen, The Netherlands 1-4 March 1982.
Hill R.L., Gordon A.J. and Neser S. (2000). The potential role of Bruchophagus acaciae (Cameron) (Hymenoptera: Eurytomidae) in the integrated control of Acacia species in South Africa. Pp. 919-929 In: Proceedings of the X International Symposium on Biological Control of Weeds, N.R. Spencer (Ed.) Montana State University, Bozeman, Montana, USA.
Hill R.L., O'Donnell D.J., Gourlay A.H. and Speed C.B. (1995). The suitability of Agonopterix ulicetella (Lepidoptera: Oecophoridae) as a biological control agent for Ulex europaeus Fabaceae: Genisteae) in New Zealand. Biocontrol Science and Technology 5: 3-10.
Laing J.E. and Heraty J.M. (1981). Establishment in Canada of the parasite Apanteles pedias Nixon on the spotted tentiform leafminer, Phyllonorycter blancardella (F.). Environmental Entomology 10: 933-935.
Memmott J., Fowler S.V., Hill R.L. (1998). The effect of release size on the probability of establishment of biological control agents: Gorse thrips (Sericothrips staphylinus) released against gorse (Ulex europaeus) in New Zealand. Biocontrol Science and Technology 8: 103-115.
van den Bosch, R., Messenger, P.S. and Gutierrez, A.P. (1982). An introduction to biological control. Intext Educational Publishers, Plenum Press, New York and London. Pp 247
Preparing for a public hearing