Historically, the APLC has relied on fenitrothion (Sumitomo Chemical, Osaka, Japan), an organophosphorus compound, as the most cost-effective pesticide to suppress locust populations. Despite fenitrothion's efficacy for locust control, concerns about its environmental impact in other countries have led to its registration being reviewed in Australia (NRA 1999). Such concerns have necessitated the APLC investigating alternative control agents for use against its target species, reviewing its criteria for control, and evaluating new approaches to monitoring for the effects of locust-control agents on the environment. Figure 1 gives the average areas sprayed and the liters of insecticide used per year by the APLC between 1977 and 2000. Fenitrothion makes up the majority of the data presented in Figure 1, however, the use of fipronil (Adonis, Rhǒne-Poulenc, now BASF, Ludwigshafen, Germany) and Metarhizium anisopliae var. acridum (see below) by the APLC has increased significantly since 1997 and 1999 respectively.
An integrated approach to locust control (Hunter 2004) has allowed the APLC to continue operating while improving its environmental performance and minimizing the impact of locust control on the environment. What follows is a brief précis of the control agents, both chemical and biological, used by the APLC and their benefits and disadvantages.
Fenitrothion and vertebrates
Fenitrothion, O,O-dimethyl O-(3-methyl-4-nitrophenol) phosphorothioate, is a broad-spectrum organophosphorus (OP) insecticide used throughout the world for the control of agricultural and forest pest and disease vectors affecting humans and animals (Bunn et al. 1993). The APLC has been using fenitrothion as its chemical of choice since it 1st began operations in 1976. During this period, there has been only 1 known incident in which birds appear to have been poisoned as a direct result of APLC operations.
During experimental spray trials in SW Queensland, in 1992, black kites (Milvus migrans Boddaert) were accidentally exposed to fenitrothion at application rates significantly higher than those currently used in APLC operations. Analysis revealed that they had been gorge feeding on C. terminifera nymphs, which had been significantly overdosed as a result of the altered spray parameters during these trials. Posttreatment checks of the site found a total of 15 dead black kites. Normal APLC spray parameters involve reduced application rates (from 381 to 267 g active ingredient [ai] • ha−1) and, therefore, a repeat of these circumstances is considered unlikely.
To quantify the pesticide doses birds are likely to be exposed to in the field, the APLC is collaborating with the University of Wollongong (New South Wales, Australia) and Texas Tech University (Lubbock, TX, USA) in a range of field- and laboratory-based research projects investigating the sublethal effects of pesticides on vertebrates. In addition to locust and vegetation residue sampling aimed at identifying the levels of bioavailability of pesticides (research that is not yet complete), birds are captured before and after spraying to measure levels of pesticide residues and OP biomarkers (cholinesterase activity) in their blood. The sublethal effects of similar low doses of fenitrothion on the metabolic (body chemistry), thermoregulatory (temperature control), and locomotory (movement) performance of a few key native species are also being studied. These performance measures provide an excellent integration of whole-animal health and can be directly related to survival in the field (Szabo et al. 2003). This research is extended across a range of vertebrate taxa.
Figure Figure 1.. Mean amounts of insecticide and areas sprayed by the Australian Plague Locust Commission (APLC) 1977 to 2000.
Download figure to PowerPoint
To date, results have demonstrated that the level of exposure of species feeding in sprayed areas can be quite high, even for birds not consuming locusts (e.g., zebra finches). Laboratory studies show that fenitrothion can have effects on locomotion (and, therefore, predator-escape capability) for much longer periods (days to weeks) than predicted from biomarker persistence in the blood (Szabo et al. 2003).
Fenitrothion generally breaks down rapidly in vertebrates and does not accumulate in body organs (Shore and Douben 1994). The indicator of exposure most widely used is cholinesterase inhibition in blood (Hamblin and Golz 1953; Rider et al. 1957; Ganlin et al. 1964; Gough et al. 1967) or brain tissue (Rider et al. 1957; Brust et al. 1971; Ludke et al. 1975; Greig-Smith 1991; Thompson 1991). Acute toxicity is generally associated with a depression in brain acetylcholinesterase activity of greater than 50 to 70%, whereas inhibitions in excess of 20%, or 2 SD below the reported norm, have been used as exposure criteria following known application of anticholinesterase pesticides (Rattner and Fairbrother 1991). Plasma cholinesterase can be inhibited to levels close to 0 without severe inhibition of brain acetylcholinesterase or overt toxic effects (Thompson 1991), and several studies have indicated that inhibition of greater than 50% in living vertebrates is not necessarily life threatening, at least in laboratory settings (e.g., Ludke et al. 1975; Evans and Batty 1986; Dell'Omo et al. 1997).
A review of the effects of OP and carbamate insecticides on vertebrate taxa found that most information on the hazard to wildlife is based on knowledge of their environmental fate, persistence, application rate, and toxicity (Story and Cox 2001). In many cases, toxicity data are gained from laboratory animals in controlled conditions. This information is often used to predict the risk to individual animals or populations in the wild. Such methods can successfully identify effects OPs have on free-living animals but have limited use when extrapolated quantitatively because differences in species sensitivity to pesticides and variation in exposure patterns between laboratory and wild animals are yet to be quantified. In addition, controlled laboratory tests can not be relied on to predict alterations in wild populations.
Organophosphorus pesticides may not only have a direct effect on individuals within a species but also populations as a whole, disrupting population dynamics and changing species composition, with the most sensitive species becoming extinct in areas of highest contamination (Story and Cox 2001). Effects of OP and carbamate pesticides can be recognized by various responses, including impaired thermoregulatory functions, changes in levels of activity and aggression, or reduced food and water intake. Hence, sublethal effects have the potential to inhibit reproduction and survival (Story and Cox 2001).
A recent pilot study to investigate the sublethal effects of fenitrothion on the metabolism and exercise performance of fat-tailed dunnarts (Sminthopsis crassicaudata [Gould]) has helped to fill an important gap in our knowledge of the effects of xenobiotics on native Australian fauna (W.A. Buttemer et al. 2005). That study also addressed the issue of inadequate linking of laboratory and field data in past studies and found that the running endurance of dunnarts (at 1 m · s−1) was reduced by up to 50% after fenitrothion (30 mg · kg−1) ingestion, although peak metabolic rate and cost of transport were unaffected during both exercise and cold exposure. Total brain cholinesterase and plasma acetylcholinesterase in these dunnarts were both reduced as a consequence of exposure to fenitrothion but had returned to approximate, predose levels by day 5 after ingestion. In addition, a consistent symptomology timeline was observed in all animals (but for varying durations) that included immobilization, excessive urination, tremors, and salivation for up to 5 h after fenitrothion ingestion (Buttemer et al. 2005).
These results indicate a potential increase in sensitivity to fenitrothion exhibited by the marsupial S. crassicaudata over that shown by similar-sized, eutherian mammals. For example, the acute, oral, fenitrothion LD50 for Mus domesticus is 240 mg · kg−1, significantly higher than the 30 mg · kg−1 dose administered to S. crassicaudata, despite being an LD50 value. It could be argued that the approach taken by Buttemer et al. (2005) gives a more precise indication of the potential effects of xenobiotics on vertebrates than do LD50 values because aerobic capacity requires the integrated function of a suite of enzymes, organelles, cells, tissues, organs, and organ systems and, as such, provides a more pertinent measure of an individual's susceptibility and reaction to any stressor. Similarly, acetylcholinesterase-reactivation techniques can remove the problem of within-population variation as each individual animal essentially acts as its own control.
Fipronil, 5-amino-3-cyano-1-(2,6-dichloro-4-trifluoromethylphenyl)-4-trifluoromethylsulfinylpyrazole, a phenyl pyrazole compound, is a broad-spectrum, low-dose insecticide that works via direct contact and stomach action. It is highly effective against both the nymph and adult stages of locusts and grasshoppers. Although not as fast acting as some other insecticides currently used for locust control, it does work at very low doses and has longer residual activity. Fipronil is an extremely active molecule and is a potent disrupter of the insect central nervous system via interference with the passage of chlorine ions through the chlorine channel regulated by γ-aminobutyric acid.
Fipronil can be applied to target areas as either a blanket or barrier treatment. Blanket treatments are used to treat small blocks of 1 to 5 km2 with doses of < 1 g ai · ha, whereas barrier treatments are used mainly against larger areas infested with mobile bands of nymphs. This method exploits the residual activity of fipronil with individual spray runs spaced widely across target areas (usually 500 m apart) leaving strips of treated vegetation (the barriers) containing approximately 0.25 to 0.5 g ai · ha−1. Bands of nymphs encountering these barriers in the 10 d following spraying come into contact with the treated vegetation and die. The advantages of this method include reduced insecticide use and aircraft costs as well as lower overall doses in target areas.
The APLC began operational use of the ultralow-volume formulation of fipronil in spring 2000. Aerial application of fipronil to control migratory locusts (L. migratoria) in Madagascar, applied at a rate of 3.2 to 4.0 g ai · ha−1, has been shown to have a long-term effect on populations of mound-building termites and their specialized vertebrate predators (Peveling et al. 2003). The APLC uses much lower application rates of fipronil (0.25–1 g ai · ha−1) to control locusts in Australia and few, if any, mound-building termites occur in sprayed areas. To determine the effect of fipronil on invertebrates and vertebrates in Australian grasslands, the APLC initiated the first of an intended series of impact studies in February 2002. In this trial, fipronil (1.25 g ai · ha−1) was aerially applied to bands of C. terminifera nymphs infesting an open, Mitchell grass (Astrebla spp.) plain in far-western Queensland. The fipronil was compared with a nearby area sprayed with fenitrothion (267 g ai · ha−1) and an untreated control. Invertebrates were monitored immediately before and after spraying using pitfall traps, yellow-pan traps, and malaise traps. Sampling is still ongoing and has been conducted at irregular intervals for more than a year.
Preliminary analysis of the results shows that both insecticides had a similar short-term impact on the number of invertebrates caught, with most groups showing recovery when sampled about 11 weeks after spraying. Full recovery of populations from perturbation by insecticides, however, has been affected by a prolonged drought. No substantial rain fell on the trial site until February 2003, 12 months after spraying. A sample taken in late March 2003 indicated that partial recovery of invertebrate populations had occurred in sprayed and unsprayed areas but not to the levels before spraying. As in previous studies, Collembola and ants were found to be the most useful indicator groups because they were caught in high enough numbers for statistical analysis and showed sensitivity to the insecticide treatments. Both groups are currently being identified to morphospecies to determine whether the insecticides have had an effect on species diversity and community structure.
Mound-building termites were not present on the trial site, but subterranean nesting species were commonly found infesting cattle dung. No effect on the percentage of cattle dung infested by these termites was detected 7 weeks after spraying, but further work on this group is needed and is planned in future trials. Trials to assess the impact of fipronil in contrasting locust habitats, such as central New South Wales rangelands, are planned when suitable conditions arise.
Recent concern has been expressed over the use of fipronil in Madagascar because of its impacts on termites and subsequent effects on specialist termite predators (Peveling et al. 2003). Results from ongoing APLC research has detected fipronil and its metabolites in caudally stored fat of S. maacroura (Gould) and S. crassicaudata up to 6 months after application in arid-zone Australian rangelands. Although the amounts discovered are minute (0.8–4 ppb fiproles), the fat stored in the tails of these insectivorous, native marsupials is used throughout traditionally dry winters as an energy source. Therefore, any compromise of this valuable resource at key points in the life cycle of either S. macroura or S. crassicaudata could prove critical for the population dynamics of the species. Subsequently, a comparison of the sublethal effects of fenitrothion and fipronil has been incorporated into our research program as a high priority.
Table Table 1.. Comparison of locust-control agents currently available to the Australian Plague Locust Commission (APLC)
| ||Fenitrothion||Fipronil (blanket treatment)||Fipronil (barrier treatment)a||Metarhiziumb||IGRs|
|Risk to wildlife|| || || || || |
| Aquatic arthropods||Moderatec||Lowd||Low||Low||High|
|Risk to human healthf||Moderately hazardous||Unlikely||Unlikely||Unlikely||Unlikely|
|WHO toxicity classf||II||Unknown||Unknown|| || |
|Mode of action||AChE inhibition||Blocks GABA receptorsg||Blocks GABA receptors||Fungal disease||Inhibits molting|
|Speed of action||Fast, 2–3 d||Moderate, 2–7 d||Moderate, 2–7 d||Slow, 7–14 d||Slow, 4–10 d|
|Withholding periods|| || || || || |
| Livestockh||14 d||21 d ESLg||21 d ESL||NDg||21 d ESL|
| Harvestingi||14 d||14 d||14 d||ND||ND|
The native fungus, M. anisopliae var. acridum (Driver and Milner) (isolate FI-985), marketed as Green Guard (Becker Underwood Pty Ltd., Somersby, New South Wales, Australia), can be applied using aerial or ground equipment for locust control. It was originally isolated from a dead spur-throated locust, A. guttulosa, in 1979 but has also been isolated from acridid grasshoppers in Africa, Asia, South America, and North America (Mexico). The APLC applies M. anisopliae at a rate of 25 spores suspended in a 500-ml mixture of mineral and corn oil per ha. Spores can either land on locusts directly during application or can be picked up as they move through vegetation (Scanlan et al. 2001). Live spores germinate when they contact the cuticle of the locust or grasshopper and then grow into the body. Once inside the target host, the fungus is usually undetectable for several days, after which, it can be found as hyphal fragments initially in the gut and later spreading to fat deposits. In the field, the host is usually killed within 1 to 2 weeks, although mortality can take 3 to 5 weeks when temperatures for fungal development are unfavorable.
Locust control in Australia is an increasingly difficult task to undertake. Constraints facing the APLC include an expanding organic agricultural industry (particularly organic beef enterprises) and raised awareness of the locations and status of rare and threatened species and ecosystems, as well as the need to maintain 1,500-m spray buffers around waterways. Green Guard provides the APLC with an opportunity to manage these situations and continue operations where, previously, the only option was total avoidance. For example, during spring 2000, with the approval of the New South Wales Parks and Wildlife Service, locust control using Green Guard was permitted in habitat considered critical to the survival of the endangered plains-wanderer (Pedionomus torquatus Gould). The plains-wanderer is currently listed as endangered in New South Wales under the Threatened Species Conservation Act (1995), and chemical locust-control agents have been identified as a threatening process for this species. Therefore, the use of Green Guard, in this instance, is the only method by which locust control can be undertaken in areas considered to be critical habitat for the plains-wanderer.
Table 1 documents the issues associated with the use of M. anisopliae var. acridum (LUBILOSA isolate IMI 330189), relative to other control agents as determined by the Food and Agriculture Organization's Pesticide Referees Group (FAO UN 1999). This group assesses locust-control agents based on available data and use patterns and makes recommendations for locust-control operations worldwide.
Figure Figure 2.. Framework for the Australian Plague Locust Commission's (APLC) Environmental Management System (EMS).
Download figure to PowerPoint
Although for some regions within the APLC's area of responsibility, there remains an acute need for the fast knockdown capabilities of insecticides, specifically, to reduce the potential for locusts to migrate to cropping zones, the use of biological control agents, such as Green Guard, is critical to maintaining the organization's effectiveness. An integrated pest management approach to locust control has been put forward by previous authors (Lomer et al. 1999; Hunter 2004), and its implementation is essential for locust-control agencies, such as the APLC, to successfully balance their core business of locust-population suppression while maintaining environmental due diligence processes, and therefore, in the case of Australia, their legislated environmental responsibilities.
Insect growth regulators
Insect growth regulators, generally the benzoylureas, are used in a number of countries to control locusts and grasshoppers. Following ingestion, insect growth regulators disrupt chitin synthesis in arthropods and prevent the insect from successfully molting to the next life stage. Typically, the old exoskeleton is shed and mortality occurs when a new exoskeleton cannot be formed. Sublethal doses can result in mortality from other causes, such as predation, when wing or limb formation in the locust is impaired. Mortality, although not immediately apparent, normally occurs between 4 and 10 d after treatment, reflecting the typical duration of a nymphal instar.
IGRs have a significant advantage over fenitrothion and fipronil in that they have very low vertebrate toxicity. They also have little direct effect on adult arthropods as these do not need to molt, thereby, allowing natural enemies to have full effect. This advantage in reduced impact on natural enemies also causes the limitation in operational use to targeting only nymphal stages of the locust. Insect growth regulators are toxic to immature stages of aquatic arthropods, and downwind buffers must be strictly imposed. Risks to trade are probably the greatest obstacle to operational use by the APLC. Their extended half-life and high lipid solubility could impact on market access for beef and lamb or mutton fed on treated pasture.