Biological Control is it environmentally safe?

Forest and Bird, Issue 282, 1 November 1996, Unnumbered Page

Biological Control is it environmentally safe?

BARBARA

BARRATT

Every year a number of alien organisms are deliberately introduced into New Zealand to control agricultural insect pests. But the assurances regarding the environmental safety of biological control are now being

questioned.

looks at some

potential problems for our native insect fauna.

MPORTING ANIMALS to control other animals has a long tradition in New Zealand — a country of many exotic introductions. One of the most infamous early attempts at biological control was the introduction of weasels, ferrets and stoats in the 1880s to reduce rabbit numbers. The attempt failed when these generalist European predators turned their attention to easier prey such as native birds. Today biological control is more refined and is a widely accepted method of controlling the pests of economic crops to improve yields. It is also being examined as a way of reducing rabbit numbers (rabbit calicivirus disease) and for the control of possums (viral immunocontraception). The first deliberate introduction of a biological control agent into New Zealand was thought to be the 11-spotted ladybird in 1874, for control of aphids. Since then a large number of organisms have been brought to this country to control insect pests and weeds. Up to 1992, 242 biological control agents — mostly from Australia, Europe and North America — had been released to control 70 target pest species. Most of the introductions have failed to establish here. Of the 75 that have established, most have failed to control their specified pest target. In fact, only 22 introduced biological control agents — less than ten percent of those released — are reported to have had some impact on the specified target species, although the success rate in New Zealand is better than the international average. Some of these introductions were aimed at native species which have become pests, such as grass grub (which increased in numbers with the develop-

ment of pastures), but most have targeted the pests of horticulture, crops, pasture and forestry that have become established in New Zealand from overseas. When a new pest arrives in New Zealand, it usually comes without the suite of natural enemies that keep it in check in its country of origin. Modern biological control is an attempt to redress the balance, by importing a natural enemy — a predator, a parasite or a disease-causing pathogen — from the area of origin of the pest. This is known as "classical biological control". Once released into the environment, these new organisms are usually left to maintain themselves and spread, reaching a balance with the pest species they are intended to control, hopefully keeping them below damaging population densities. Finding a potential biological control agent for a new pest usually requires exploration and study in the pest’s area of origin, but sometimes a control agent can be imported from another country where the same pest has established, and a suitable biological control agent has already been found. Another type of biological control is known as "inundative control", which is used, for example, in glasshouse crops in New Zealand for whitefly and mites. Growers can purchase mass-produced biological control agents for these pests which they liberate into the glasshouse to give control for a season, or for the growing period of the crop. This needs to be repeated in subsequent years.

IOLOGICAL CONTROL is one of ie many tools used in weed and pest management. After the 1962 publication of Rachel Carson’s Silent Spring attacking the widespread use of pesticides, the public became increasingly aware of the potential dangers to the environment, to wildlife and to themselves from largescale pesticide use. The effect was to force the hand of regulatory bodies around the world to review their pest control strategies, which until then relied heavily on the use of chemical herbicides and insecticides. Reductions in pesticide use can be achieved by implementing a system which combines methods of weed and pest control known as "integrated pest management". Chemical pesticides still have a role in integrated systems, but the emphasis has steadily shifted towards combinations of improved husbandry, and biological control. Amongst the strong environmental arguments raised against the use of pesticides are the lasting residual effects (the time it takes chemicals to break down in the environment) and the effects on nontarget wildlife which occur during the passage of these materials through the food chain. Many of the worst offenders, such as DDT and dieldrin (organochlorines), have generally been banned from use, although they still remain a problem in the environment. Biological control, by comparison, has been enthusiastically promoted as a safe and natural alternative to pesticides, and

The approval process

_-@ NDER THE HAZARDOUS Substances And New Organisms Act passed earlier this year, an Environmental Risk Management Authority (ERMA) will be established to review applications to import new organisms including biological control agents. This will replace the current arrangement where applications are received by the Ministry of Agriculture’s Regulatory Authority, and vetted (including a public review process) before an import permit is issued by the Chief Veterinary Officer. At this stage it is expected that a similar process will be followed by ERMA. The public, as well as major stakeholders such as DoC and Crown Research Institutes will be invited to make submissions on applications.

Minimum standards will have to be met and applications will be declined if the new organism is likely to significantly displace native species, cause deterioration of natural habitats, or harm New Zealand’s inherent genetic diversity. Any new organisms imported will also require biosecurity and import health clearance from MAF under the 1993 Biosecurity Act before release. One of the short-comings of this process is that there are currently no guidelines for testing to ensure that biological control agents will only attack the targeted pest, or protocols which can be implemented to ascertain that the minimum standards are met. There is also no requirement for postrelease monitoring of new organisms released into the environment.

conservationists a. have generally welcomed the move towards what appears to be a more environmentally acceptable form of pest management. There have been some well-publicised and spectacular successes in biological control. One example is the control of prickly pear cactus in Australia after the introduction from Argentina of the Cactoblastis moth. Another success was the release of the predatory ladybird Rodolia cardinalis from Australia into California to control cottony cushion scale on citrus crops. Although the failure rate is

high, biological control has in many instances undoubtedly proved to be a highly effective form of agricultural pest management. In comparison with chemical control, biological control is characterised by being irreversible, self-perpetuating and selfdispersing. In other words, when a biological control agent is released into the environment, it is done so with the expectation that it will be a permanent addition to the biota, it will reproduce and increase

in numbers from the original numbers released, and that it will spread from where it was released. These attributes are, of course, some of the very features that are seen as advantages for biological control as part of cost-effective, sustainable, pest management programmes. However, they are also factors that are beginning to alert researchers to the potential environmental implications of such introductions. In recent years the assurances regarding the environmental safety of biological control have been increasingly questioned. Information and examples have come forward on "non-target" effects, where the biological control agent is attacking species other than the one it was intended to control. In a few extreme cases in other countries, extinctions of non-target species are thought to have occurred, sometimes as a result of "hostswitching" by the agent after it has reduced numbers of the intended host to low levels. Information on these environmental impacts is scant, probably because very little research has been carried out to look for these effects. The great majority of control agents introduced to this country have been insects and other invertebrates. Thus any non-target impacts are likely to be on native (and other established) invertebrates, or possibly on native vegetation. Although no known extinctions have resulted from biological control introductions in New Zealand, some are considered to have adversely affected the native fauna. In the 1960s the parasitic tachinid fly

Native broadnosed weevils

DULT WEEVILS are characterised by the presence of a rostrum, or an elongation of the front of the head which bears the mouthparts. In broad-nosed weevils this is short and blunt. There are some 1,500 endemic species of weevils in New Zealand, and the broad-nosed weevils (Brachycerinae) are a large sub-group of these.

The native brachycerine weevil fauna is very poorly known, and the current research programme at Invermay into the possible impacts of two introduced weevils has brought a number of new species to light. In particular, little is known about the biology and ecology of these native weevils. The adults feed on seedling plants and mature plant foliage of a broad range of plants. Field observations suggest that they also feed on pollen of flowering plants, and adult survival studies in the laboratory have indicated that pollen may represent an important source of protein for egg production. Larvae are soil dwelling and feed on plant roots. The adults are generally flightless, and some species have very limited distributions. However, little is known of their conservation needs.

Trigonospila brevifacies was released for control of light brown apple moth, a pest in orchards. It is now found parasitising a number of other moth species, with anecdotal evidence to suggest an associated decline in the numbers of some native leafrollers. A research programme has recently begun to attempt to quantify this impact. Vanessa Munro working with the Horticultural Research Institute has found that native moths are parasitised in some native habitats, but the range and susceptibility of species are yet to be determined. Also, numbers of the native red admiral butterfly are thought by some entomologists to have been reduced since the introduction of parasites to control the cabbage white butterfly. T THE INVERMAY Agricultural Centre in Mosgiel, a Biological Control Group is conducting research aimed at improving the environmental safety of biological control agents introduced into New Zealand. The plan is to develop guidelines and protocols to test for host specificity — that is to identify the likely host range so that informed decisions on the impact of releases can be made — while the new organisms are still in quarantine. To do this, research

is being ora carried out on two biological control agents already introduced into New Zealand, to find out what native species they attack in the laboratory, and compare this with what is happening in the environment. The study is therefore mimicking what could be done in quarantine, while at the same time verifying the results by finding out what has actually happened in the field since they were released. The particular biological control agents chosen for

this study are two parasitic wasps. Microctonus aethiopoides was released in 1982 to control the exotic sitona weevil, a pest of lucerne, while M. hyperodae was introduced more recently, in 1991, to control the ryegrass pest, Argentine stem weevil. The wasps attack the adult stage of the weevils, laying an egg inside the body cavity. This hatches into a larva which grows as it feeds on the tissues inside the weevil, but it does not kill its host until it is ready to emerge to pupate. The weevil appears to live quite normally, BRIAN

except that females become sterile and unable to produce eggs once they are parasitised. When the parasite is fully developed it leaves the weevil, killing its host, and forms a pupa from which a new wasp emerges. The first step in our research was to identify native species that could potentially be at risk from these introduced insects. These were considered most likely to be taxonomically related native weevils, and particularly those that live in environments similar to those of the intended hosts, and would therefore be recognised by the parasitic wasps as possible targets. Both target pests — sitona weevil and Argentine stem weevil — are members of the large broad-nosed weevil subfamily, Brachycerinae, a group that includes many native species in New Zealand. Many of these native weevils are found in modified pastures as well as their natural tussock grassland and alpine environments. Their distribution therefore overlaps with the pest species, especially in pastures and semi-modified grassland areas, but both sitona weevil and the Argentine stem weevil can also be found occasionally in native grasslands, extending up to the alpine zone. From a list of related native species collected during surveys of grassland, a number were selected for laboratory tests, in which they were held in cages and exposed to the parasitic wasps. The results of these tests indicated that one of the introduced wasp species, M. aethiopoides, was very much more successful in attacking and developing in native weevils, than the other — both in terms of the number of species in which the wasps laid eggs, and the number of individual weevils in each test that were parasitised (see table

above). Most of the native species parasitised by M. aethiopoides are in the genera Irenimus and Nicaeana, but one was the subalpine to alpine species Zenagraphus metallescens, a weevil that is two to three times the size of the intended sitona host. To see how well laboratory tests could "predict" what might happen in the field, surveys of native weevils have been carried out to ascertain whether they are being parasitised "naturally" in the environment. From these studies, we found that again, M. aethiopoides is far more successful in exploiting native weevils as hosts, than is M. hyperodae, with up to 70 percent of one native weevil population in pasture being attacked by the former. While the result was encouraging in indicating that laboratory testing can give a useful indication of likely impact in the environment, it was also worrying that such a non-specific parasitic organism has

been released. The studies will continue with regular monitoring of some of the native weevil species that are being parasitised, so that we can gain an understanding of longer-term effects of M. aethiopoides on the species concerned. IOLOGICAL CONTROL agents are usually released to combat an agricultural pest problem, but obviously once they are released, they spread by their own means into any suitable environment including conservation areas. The research at Invermay has shown that introduced biological control agents can pose a risk to native species, although more work is required to work out the extent to which native weevil populations are threatened, and whether other introduced biological control agents are having a similar impact. The environmental implications of non-target effects of biological control agents are clearly complex, ranging from direct effects upon the survival of non-tar-get hosts, to ecological ramifications which follow in food webs when there is a substantial change to the status of any species in an ecosystem. Depending upon the role of the "at risk" species in the ecosystem, and the balance and complexity of the system itself, changes may be almost insignificant, or conversely they may impact severely upon a number of other species. This makes prediction of effects extremely difficult. It is important to realise also that effects might not be noticed for many years. The chances of an introduced biological agent affecting non-target organisms increases over time as the introduced agent spreads and comes into contact with more native species. Our research programme deals with

Parasitism of native weevils

Although it was reassuring to find the laboratory tests matching results in the field, it was worrying to discover what was thought to be a Wasps introduced as biological control agents reasonably host-specific biological control agent | Microctonus Microctonus attacking a number of native weevil species. aethiopoides hyperodae Laboratory tests No. native species parasitised/no. tested 7/7 5/7 Average parasitism 58% 13% Field monitoring Number native species parasitised 13 Maximum parasitism recorded 71% 3% Number of sites where parasitism was found in native species 10 l

biological control agents introduced for insect control. Biological control of weeds presents a similar threat to the environment as there is always a danger of the new organism switching to native plants. However, there are significant differences. In New Zealand we have approximately 2,000 species of native plants of which probably 95 percent have been described. This means that, given the resources, scientists can thoroughly test the new organism to see if it will attack related native plant species (and economically important plants), and thus be more confident in assessing the degree of risk involved in the release of the biological control agent. For insects the situation is very different. There are an estimated 20,000 species of native insects, of which about half are described, and a large number yet to be discovered. This makes quarantine testing much more difficult, and emphasises the need for urgent taxonomic work to be carried out on the New Zealand insect fauna. Biological control is a powerful pest management tool with both environmental and economic risks and benefits that need to be considered. It is often argued that the costs of carrying out detailed quarantine investigations to assess potential environmental impacts are not warranted. However, it can also be argued that the irreversible addition of a new organism to the environment at least justifies the cost of a minimum level of investigation so that a realistic analysis of the level of risk to the environment can be ascertained. The new procedures to be developed under the Hazardous Substances and

New Organisms Act will hopefully make this mandatory. When M. aethiopoides was released in New Zealand, it was done so with very little quarantine testing. It has been reasonably successful in controlling its intended host, but time will tell as to whether the benefits will be outweighed by the environmental costs. As an interesting twist to the story, which makes the cost-benefit equation even more complex, M. aethiopoides has now been found parasitising a beneficial weed biological control agent, the nodding thistle receptacle weevil (Rhinocyllus conicus), and could therefore be compromising the benefits of another biological control programme. The more recent introduction of this problematic wasp’s cousin, M. hyperodae, did not occur until after about a year of careful investigation in quarantine of potential host range, and extensive peer review of the findings. It was known that some native species could be attacked in the laboratory, but the research indicated that only one native species was likely to be attacked in the field. Although it is only five years after its release, only one native species has been shown to be attacked in the field, and it is the one predicted from quarantine studies. @

Dr BarBARA BARRATT is a scientist with AgResearch in the Biological Control Group based at Invermay Agricul- _ tural Centre, Mosgiel.