R.J. DouthwaiteThe author is head of the Field Ecology Section, Natural Resources
Institute, Central Avenue, Chatham Maritime, ME4 4TB, UK.
Experiments with insecticides for tsetse fly control began in 1945 and, two years later, resulted in a full-scale campaign to eradicate Glossina pallidipes, G. brevipalpis and G. austeni from their last foci in Zululand, South Africa (Du Toit, 1954). Persistent organochlorine insecticides were dispersed indiscriminately in dust or smoke over approximately 18 000 km2 of bush both by air and from the ground. The successful eradication of the fly was achieved in 1954, by which time many areas had received eight treatments of DDT (approximately 400 g a.i./ha) or of BHC (approximately 80 g a.i./ha). The effects on non-target wildlife can only be guessed.
Concern was expressed for the fate of beneficial insects (Du Toit, 1954) but the long-term hazards of organochlorine residues, because of their persistence and bioaccumulation, were not recognized at the time. The use of insecticides removed the threat of further game slaughtering in the Umfolozi, Hluhluwe and Mkuzi game reserves where 138 000 animals had been killed up to 1946 in an attempt to starve the tsetse out. The legacy of insecticide use may have been the virtual extinction of epiphytic orchids and other plants that are solely dependent on a few species of co-evolved hymenopterans for pollination (Downing and Gibbs Russell, 1981).
The widespread use of insecticides for tsetse fly control followed and, over the years, different methods evolved in response to economic and environmental pressures. The indiscriminate spraying of high doses, sufficient to persist for three or four months, has largely been replaced by selective, residual treatment of tsetse resting sites or by sequential, indiscriminate low-dose drift sprays. Treatment is timed to kill newly emerged flies before they deposit larvae on the ground and is repeated until no more flies are left to emerge from pupae under the ground. More recently, visual and olfactory lures, selected to attract tsetse flies to insecticide-impregnated cloth, have increasingly replaced spraying as the preferred means of control. The implications of these changes for non-target wildlife are reviewed here.
Residual spraying
Concern for the fate of non-target insects (Du Toit, 1954) and warnings about the dangers of organochlorine insecticides (Carson, 1962) were largely ignored in the early determination to control tsetse flies with insecticides. More than 300 000 km2 of Africa has now been treated with dieldrin, DDT and gamma-BHC (Matthiessen and Douthwaite, 1985).
Several studies show that monkeys, antelopes, dogs, bats, squirrels, field mice, birds, lizards, snakes, amphibia, fish and invertebrates can be killed by the indiscriminate spraying of dieldrin, applied at rates of 680 to 1 000 g/ha from helicopters and mist-blowers (Graham, 1964; Koeman and Pennings, 1970; Koeman et al., 1971, 1978; Muller, Nagel and Flacke, 1981; Wilson, 1972), but only two studies continued for more than a few weeks to assess the longer-term impact on populations. In Nigeria two insectivorous bird species were absent from the study area a year after treatment and populations of other species had not fully recovered (Koeman et al., 1971). No irreversible effects were detected after spraying in Cameroon but arthropod numbers were reduced, some butterfly species were absent three years later and a common shrew, Crocidura denti, only reappeared four years later (Muller, personal communication, 1983).
Photo/Foto: R. Allsopp
Deltamethrin and other synthetic pyrethroids, applied indiscriminately from mist-blowers and helicopters, have replaced dieldrin in West Africa in recent years (Everts, 1979; Everts et al., 1983; Smies et al., 1980). Deltamethrin, applied at dose rates of 12.5 to 30 g/ha, had obvious impacts on terrestrial and aquatic arthropods and shrimps, and mayfly larvae disappeared for up to a year after treatment. Few effects on vertebrates have been noted although fish, birds and galagos, Galago senegalensis, may have died in Cameroon after an application of 30 g/ha (Muller, personal communication). The ecological consequences of heavy invertebrate mortality for the functioning of aquatic ecosystems and for vertebrates have not been studied.
Selective knapsack-spraying of DDT on tsetse resting sites was found to be effective in Kenya as early as 1951 (Grover, Le Roux and Parker, 1960). The technique was further developed in Zimbabwe where applications of DDT at approximately 230 g/ha since 1967 have cleared the tsetse from about 40 000 km2 of woodland (FAO, 1989). Annual treatments over periods of one to six years have usually been carried out to eradicate flies, but up to 14 applications have been necessary in some areas. Despite longstanding fears that serious side-effects would ensue, the impact on wildlife has only recently been investigated.
A major study of the impact of DDT on wildlife in tsetse control areas was conducted between 1987 and 1990. Generally, DDT residues are dissipated rapidly from tree bark and soil and there is no year-to-year accumulation, although localized "hot spots" may persist for more than a year (Matthiessen, 1985; Grant, personal communication, 1992). Residues do not affect rates of soil nitrification or respiration, nor the rate of leaf litter decomposition from small mesh litter bags, but leaf litter disappeared more slowly from bags admitting macro-invertebrates in a sprayed area than from similar bags in an unsprayed area. No major differences were detected in the epigeal arthropod faunas of sprayed and unsprayed areas. However, high residue levels accumulated along some food chains and populations of two common woodland birds, the red-billed wood-hoopoe (Phoeniculus purpureus) and white-headed black chat (Thamnolaea arnoti), which both feed on arthropods on tree trunks, and the African goshawk (Accipiter tachiro), which eats birds, were almost eradicated in sprayed areas. Their populations are unlikely to recover fully for perhaps ten years or more after spraying. Numbers of most other woodland bird species were unaffected by spraying, varying more on a year-to-year basis than as a result of direct spray treatment.
The common arboreal lizard, Mabuya striata, was less common in sprayed woodland than in unsprayed woodland. Residue levels in bats gave no cause for concern among small species although some individuals from samples of the larger species (Nycteris thebaica, Rhinolophus hildebrandtii and Scotophilus bourbonicus) retained enough DDT to approach lethal levels during periods of body stress. The level of foraging activity and reproductive success for bats in sprayed areas generally did not differ from that in untreated sites and there was no decline in species diversity in sprayed areas. However, the results indicate that potential roosting sites such as tree hollows and caves should be sprayed with alternative insecticides. There was no evidence to show that the abundance or survival of fish were affected, but high levels of DDT, linked with recent tsetse spraying operations, are causing eggshell thinning and hatching failure in fish eagles at the eastern end of Lake Kariba. Despite widespread hatching failure at the eastern end, the breeding population is as dense as it is anywhere on the Zimbabwean shore. The results of the study suggest adverse impacts are less widespread than was feared but they are sufficiently serious to warrant the use of alternative means of tsetse control wherever possible.
Non-target impacts of deltamethrin, a potential substitute to DDT for selective, residual treatments, have also been investigated recently in Zimbabwe (Lambert et al., 1991). The insecticide was applied at the rate of approximately 2.5 g a.i./ha to tsetse resting sites. It proved acutely toxic to a wide range of invertebrates found on tree trunks, and some populations, notably plant-hoppers (Homoptera: Fulgoroidea) and silverfish (Thysanura: Lepismatidae), remained significantly reduced after two months. However, no effects were found on a population of the arborial lizard, Mabuya striata. Spraying also increased the rate of downstream drift of aquatic invertebrates for several days. Mayflies (Ephemeroptera) were particularly affected but residual populations were not seriously impoverished. Seasonal drying of streams in the area was more significant than spraying operations in limiting populations. Generally, deltamethrin treatments affect invertebrates more than DDT does, but residues do not persist long or bioaccumulate, and there can be little doubt that it will prove less hazardous for vertebrates.
Sequential spray treatments
Interest in aerial spraying resumed in 1968 when endosulfan was applied in low doses (28 g/ha) in thermal exhaust to eradicate tsetse flies in western Zambia (Park et al., 1972; Magadza, 1978). Spray cycles were timed to kill newly emerged flies before they produced larvae and were repeated until no more flies were left to emerge from pupae in the ground. The success of the operation awakened wider interest in aerial spraying and drift sprays of endosulfan have now been applied to more than 100 000 km2 throughout Africa. Smaller areas have also been treated with deltamethrin or with other synthetic pyrethroids and pyrethroid/endosulfan mixtures.
The impact of endosulfan drift sprays, applied from aircraft at dose rates of 6 to 41 g a.i./ha, has been studied in Botswana, Zimbabwe, Zambia, Somalia and Burkina Faso. Side-effects mainly arose when navigational errors or leaking equipment caused local overdosing. Fish, especially small ones found in clear, shallow water, are most susceptible. In Botswana spray cycles of 6 to 12 g endosulfan/ha killed from 0 to 4 percent of the fish population, varying according to species (Fox and Matthiessen, 1982). Breeding among Tilapia rendalli was disrupted and tissue damage to the brain and liver was found in survivors of several species (Matthiessen and Roberts, 1982). However the impact on the fish population was less than that caused by the seasonal drying of rivers.
Increased mortality, hyperactivity and hive desertion were found in honey-bees (Douthwaite, Aye and Ibrahim, 1988) but, remarkably, no other effects in invertebrates have been found. Abnormally high and low feeding rates have been seen in little bee-eaters (Merops pusillus) and pied kingfishers (Ceryle rudis) after some spray cycles, and starvation for 24 hrs caused breeding failure among little bee-eaters after one spray cycle, but population effects were absent (Douthwaite, 1982, 1986; Douthwaite and Fry, 1982).
Effects of sequential drift sprays of deltamethrin have been studied in several countries. Deltamethrin is a broad-spectrum insecticide and drift sprays of 0.1 to 0.25 g a.i./ha "knock down" a wide range of arthropods in large numbers. In Botswana trees subject to one and four spray cycles held 75 percent and 18 percent, respectively, of the standing crop of arthropods in unsprayed trees (Davies, personal communication; Games, 1981). In the absence of predation, Grant and Crick (1987) estimated that 80 to 90 percent of the insects knocked down recovered to fly or crawl away, but predation of affected insects may be intense. Holloway (1990) found that approximately 90 percent of moribund tsetse flies were preyed on by ants within 2.5 furs. Sprays of 0.25 g a.i./ha in Zimbabwe (Grant and Crick, 1987; Ertz, 1988) increased the rates of downstream drift in a wide range of aquatic invertebrates, but effects were transient and no population declines resulted. In Burkina Faso an application of 0.19 g/ha had little effect on catches of decapod crustaceans, but many were paralysed by treatment with 0.36 g/ha (Takken et al., 1978). No effects on fish or honey-bees have been detected, but the impact on insectivorous vertebrates has not been studied.
Visual and olfactory lures
The development of insecticide-impregnated targets, incorporating visual and olfactory tsetse attractants, has reduced application rates to as little as 0.02 g a.i. deltamethrin/ha/yr, i.e. one thousandth of the amount used in indiscriminate residual treatments. Non-target effects appear to be minimal although the lures attract other flies, notably tabanids and muscoids (Holloway and Phelps, 1991; Vale et al., 1988). Sometimes, more concern is expressed about the necessity for access roads to remote areas and for the deployment and service of targets than about the possible impacts of insecticide use. The tracks card allow easier access for poachers and settlers and, if poorly constructed, may lead to soil erosion.
History may judge that tsetse control campaigns, including the slaughter of 1.5 million head of game (Matthiessen and Douthwaite, 1985), had an ephemeral impact on the fauna of Africa but that lasting changes were a result of the rise in human population and the associated inroads of subsistence farmers on natural and semi-natural habitats. The flora and fauna of Africa's protected reserves have been more affected by a lack of investment and sound management to control fire, elephant populations, poaching and tourism than by the incidental effects of tsetse control operations. Yet concern about the use of insecticides for tsetse fly control remains.
The perception of risk is a complex process and it is paradoxical that the greatest concern in the present case is expressed by those who are most removed from tsetse control operations. The origins of this concern lie in myth and misconception. The African bush is seen by many as a special kind of Eden, rather than a complex and changing environment in which people have to live (Anderson and Grove, 1987; Graham, 1973). Society's use of the African bush, and modern technology in the form of insecticides, can only harm that ideal. Indeed, insecticides have become a "worry bead" of the affluent world. Their known adverse effects are perceived as the "tip of the iceberg" and their benefits are discounted (Kates, 1978; Marco, Hollingworth and Durham, 1987). The price of these irrational fears is that tsetse control authorities must now accept that land-use plans - which justify the loss of Eden - environmental monitoring and sensitivity in the choice of control technology are necessary if their plans are to be accepted by the public.
Anderson, D. & Grove, R. 1987. Conservation in Africa: people, policies and practice. Cambridge, Cambridge Univ Press.
Carson, R. 1962. The silent spring. Cambridge, MA, USA, Houghton Mifflin.
Douthwaite, R.J. 1982. Changes in pied kingfisher (Ceryle rudis) feeding related to endosulfan pollution from tsetse fly control operations in the Okavango Delta, Botswana. J. Appl. Ecol., 19: 133-141.
Douthwaite, R.J. 1986. Effects of drift sprays of endosulfan, applied for tsetse fly control, on breeding little bee-eaters in Somalia. Environ. Pollut. Ser. A, 41: 11-22.
Douthwaite, R.J., Aye, M.D. & Ibrahim, A.S. 1988. Effects of drift sprays of endosulfan, applied for tsetse fly control on honey-bees (Apis mellifera L.) in Somalia. J. Apic. Res., 27: 40-48.
Douthwaite, R.J. & Fry, C.H. 1982. Food and feeding behaviour of the little bee-eater, Merops pusillus in relation to tsetse fly control by insecticides. Biol. Conserv., 23: 71-78.
Downing, B.H. & Gibbs Russell, G.E. 1981. Phytogeographic and biotic relationships of a savanna in southern Africa: analysis of an angiosperm check-list from Acacia woodland in Zululand. J S. Afr. Bot., 47: 721-742.
Du Toit, R. 1954. Trypanosomiasis in Zululand and the control of tsetse flies by chemical means. Onderstepoort J. Vet. Res., 26: 317-387.
Ertz, T. 1988. The effect of ultra-low volume aerial applications of deltamethrin applied for tsetse fly control on populations of non-target organisms in northern Zimbabwe. Report to the Regional Tsetse and Trypanosomiasis Control Programme Harare, Scientific Environmental Monitoring Group.
Everts, J.W., ed. 1979 Side-effects of aerial insecticide applications against tsetse flies near Bouafle, Ivory Coast. Wageningen, Department of Toxicology, Agricultural Univ.
Everts, J.W., van Frankenhuyzen, K., Roman, B. & Koeman, J.H. 1983. Side-effects of experimental pyrethroid applications for the control of tsetse flies in a riverine forest habitat (Africa). Arch. Environ. Contam. Toxicol., 12: 91-97
FAO. 1989. Integrated tsetse control and rural development. Report of a joint meeting of the FAO panel of experts on technical, ecological and development aspects of the programme for the control of African animal trypanosomiasis and related development (7-9 November 1988, Accra, Ghana) Rome, FAO.
Fox, P.J. & Matthiessen, P. 1982. Acute toxicity to fish of low-dose aerosol applications of endosulfan to control tsetse fly in the Okavango Delta, Botswana. Environ. Pollut. Ser. A, 27: 129-142.
Games, I.P. 1981. Report on the effects of a deltamethrin and endosulfan mixture on the non-target arthropods during the 1981 tsetse spraying program. Maun, Botswana Department of Veterinary Services and Tsetse Fly Control.
Glover, P.E., Le Roux, J.G. & Parker, D.F. 1960. The extermination of Glossina palpalis on the Kuja-Migori river systems with the use of insecticides. In Proc. Int. Sci. Comm. Trypanosomiasis Res. Seventh Meeting Brussels, 1958, p. 331-342. Publ. Comm. Tech. Coop. Afr. S. Sahara, No. 41 (London).
Graham, A. 1973. The gardeners of Eden. London, George Allen & Unwin.
Graham, P. 1964. Destruction of birds and other wildlife by dieldrex spraying against tsetse fly in Bechuanaland. Arnoldia (Rhod), 1: 1-4.
Grant, I.F. & Crick, H.Q.P. 1987. Environmental impact of sequential applications of deltamethrin aerosols applied for tsetse control in Zimbabwe. London Tropical Development and Research Institute.
Holloway, M.T.P. 1990. Alternatives to DDT for use in ground spraying control operations against tsetse flies (Diptera: Glossinidae). Trans. Zimbabwe Sci. Assoc., 64: 33-40.
Holloway, M.T.P. & Phelps, R.J. 1991. The responses of Stomoxys spp. (Diptera: Muscidae) to traps and artificial host odours in the field. Bull. Entomol. Res., 81: 51-55.
Kates, R.W. 1978. Risk assessment of environmental hazard. Chichester, John Wiley & Sons.
Koeman, J.H. & Pennings, J.H. 1970. An orientational survey on the side-effects and environmental distribution of insecticides used in tsetse control in Africa. Bull. Environ. Contam. Toxicol., 5: 164-170.
Koeman, J.H., Rijksen, H.D., Smies, M., Na'Isa, B.K. & Maclennan, K.J.R. 1971. Faunal changes in a swamp habitat in Nigeria sprayed with insecticide to exterminate Glossina. Neth. J. Zool., 21: 434-463.
Koeman, J.H., Den Boer, W.M.J., Feith, A.F., de Iongh, H.H., Spliethoff, P.C., Na'Isa, B.K. & Spielberger, U. 1978. Three years' observation on side-effects of helicopter applications of insecticides used to exterminate Glossina species in Nigeria. Environ. Pollut. Ser. A, 15: 31-59.
Lambert, M.R.K., Grant, I.F., Smith, C.L., Tingle, C.C.D. & Douthwaite, R.J. 1991. Effects of deltamethrin ground-spraying on non-target wildlife. Chatham, UK, Natural Resources Institute.
Magadza, C.H.D. 1978. Field observations on the environmental effect of large-scale aerial applications of endosulfan in the eradication of Glossina morsitans centralis Westw. in the western province of Zambia in 1968 Rhod. J. Agric. Res., 16: 211-220.
Marco, G.J., Hollingworth, R.M. & Durham, W. 1987. Silent spring revisited. Washington, D.C., American Chemical Society.
Matthiessen, P. 1985. Contamination of wildlife with DDT insecticide residues in relation to tsetse fly control operations in Zimbabwe. Environ. Pollut. Ser. B, 10: 189-211.
Matthiessen, P. & Douthwaite, R.J. 1985. The impact of tsetse fly control campaigns on African wildlife. Oryx, 19: 202-209.
Matthiessen, P. & Roberts, R.J. 1982. Histopathological changes in the liver and brain of fish exposed to endosulfan insecticide during tsetse fly control operations in Botswana. J. Fish Dis., 5: 153-159.
Muller, P., Nagel, P. & Flacke, W. 1981. Ecological side effects of dieldrin application against tsetse flies in Adamaoua, Cameroon. Oecologia (Berl.), 50: 187-194.
Park, P.O., Gledhill, J.A., Alsop, N.J. & Lee, C.W. 1972. A large scale scheme for the eradication of Glossina morsitans morsitans Westw. in the Western Province of Zambia by aerial ultra-low volume applications of endosulfan. Bull. Ent. Res., 61: 373-384.
Smies, M., Evers, R.H.J., Peijnenburg, F.H.M. & Koeman, J.H. 1980. Environmental aspects of field trials with pyrethroids to eradicate tsetse fly in Nigeria. Ecotoxicol. Environ. Safety, 4: 114-128.
Takken, W., Balk, F., Jansen, R.C. & Koeman, J.H. 1978. The experimental application of insecticides from a helicopter for the control of riverine populations of Glossina tachinoides in West Africa. VI. Observations on side-effects. PANS, 24: 455-466.
Vale, G.A., Lovemore, D.F., Flint, S. & Cockbill, G.F. 1988. Odour-baited targets to control tsetse flies, Glossina spp. (Diptera: Glossinidae), in Zimbabwe. Bull. Ent. Res., 78: 31 49.
Wilson, V.J. 1972. Observations on the effect of dieldrin on wildlife during tsetse fly, Glossina morsitans, control operations in eastern Zambia. Arnoldia (Rhod.), 5: 1-12.
