The transfer of appropriate technology; key to the successful biological control of five aquatic weeds in Africa

Transcription

1 The transfer of appropriate technology; key to the successful biological control of five aquatic weeds in Africa Martin P. Hill 1 and Mic H. Julien 2 Summary The rivers and lakes of Africa have been subjected to invasion by alien aquatic vegetation since the early 1900s. Water hyacinth (Eichhornia crassipes), water lettuce (Pistia stratiotes), salvinia (Salivinia molesta), parrot s feather (Myriophyllum aquaticum) and red water fern (Azolla filiculoides), all native to South America, have been recorded in many countries in Africa. Vast mats of these weeds impact on all aspects of water utilisation and severely degrade aquatic biodiversity. Biological control programs have been initiated against all of these weeds from the early 1980s onwards and they have been successfully brought under control in many areas, although in some areas water hyacinth remains problematic. The successful biological control of aquatic weeds in Africa has been ascribed to several factors, including: the reliance on fundamental research performed on the target biocontrol agents in developed countries obviating the need to screen agents for host specificity in resource-poor countries; the development of simple mass-rearing techniques, ensuring the release of high numbers of healthy insects; and a standard post-release monitoring technique allowing comparison between different control sites. However, the single most important factor contributing to the success of these programs was the involvement of dedicated individuals who understood the potential of biological control and who ensured that the projects progressed. These people developed community involvement, another important factor in the success of projects, in mass rearing and agent distribution. Keywords: biological control, integrated control, technology transfer, water hyacinth. Introduction There are five important aquatic plant species in Africa, which warrant control: Azolla filiculoides Lamarck (Azollaceae) (red water fern); Myriophyllum aquaticum (Vellozo Conceição) Verdcourt (parrot s feather); Salvinia molesta D.S. Mitchell (Salviniaceae) (salvinia); Pistia stratiotes Linnaeus (Araceae) (water lettuce) and Eichhornia crassipes (Martius) Solms- Laubach (Pontederiaceae) (water hyacinth). The native ranges for all five of these species is South America, but they have all been introduced to many parts of the world, where they have become problematic, especially 1 Department of Zoology and Entomology, Rhodes University, PO Box 94, Grahamstown 6140, South Africa. 2 CSIRO Entomology, 120 Meiers Road, Indooroopilly, Queensland 4068, Australia. Corresponding author: M. Hill, Department of Zoology and Entomology, Rhodes University, PO Box 94, Grahamstown, 6140, South Africa <m.p.hill@ru.ac.za>. in tropical and subtropical regions (Holm et al. 1977). Two issues contribute to their invasiveness the lack of co-evolved natural enemies in their adventive range (Buckingham 1994) and the presence of nitrate and phosphate enriched waters associated with urban, agricultural and industrial pollution (Heard & Winterton 2000). These five species form dense mats on rivers and dams throughout Africa and degrade aquatic ecosystems and limit their utilisation. The economic and environmental losses due to these weeds are huge, to the extent that the problem threatened the development of Africa. One of the key issues that separates aquatic weed from terrestrial weed programs is that the impact of water weeds on riverine communities is easy to quantify which makes it easier to generate funding, unlike many of the terrestrial weeds whose impacts, notably on biodiversity, are more difficult to quantify. Here we report on the successful biological control of these five weeds in Africa. 370

2 Technology transfer of aquatic weed biocontrol Azolla filiculoides (red water fern) Azolla filiculoides has become a weed of small dams and slow-moving rivers in a number of countries around the world. The first record of this plant in Africa was from South Africa in 1948, where it was introduced as a pond ornamental (Oosthuizen & Walters 1961). Phosphate-enriched waters, the lack of natural enemies (Hill 1998a) and dispersal between water bodies by man and waterfowl facilitated an increase in its distribution and abundance. Azolla filiculoides is able to undergo rapid vegetative reproduction throughout the year by elongation and fragmentation of small fronds, and under ideal conditions, the daily rate of increase can exceed 15%. This translates to a doubling time for the weed of 5 7 days (Lumpkin & Plucknett 1982). In addition, the fern can reproduce sexually via the production of spores, especially when the plant is stressed, which can overwinter and are resistant to extreme desiccation, allowing reestablishment of the fern after drought (Ashton 1992). Among the major consequences of the dense mats (5 30 cm thick) of the weed on still and slow-moving water bodies are: reduced quality of drinking water caused by bad odour, colour and turbidity; an increase in waterborne, water-based and water-related diseases; increased siltation of rivers and dams; reduced water surface area for recreation (fishing, swimming and water-skiing) and water transport; deterioration of aquatic biological diversity (Gratwicke & Marshall 2001); clogging of irrigation pumps; drowning of livestock; and reduced water flow in irrigation canals. Mechanical and herbicide control options have been suggested for A. filiculoides. However, the impractical nature of mechanical control and undesirability of herbicide control in the aquatic environment suggested that A. filiculoides was a suitable candidate for biological control. The weevil Stenopelmus rufinasus Gyllenhal (Curculionidae) was prioritised as a biological control agent for A. filiculoides. This frond-feeding weevil was imported to South Africa from Florida (USA) in late 1995, underwent host-specificity testing in quarantine (Hill 1998b) and was cleared for release in late 1997 (Hill 1999a). The weevil has been released at some 110 sites throughout South Africa. The information available on these sites is that the weevil had been responsible for clearing 72 of them completely, all within a year (McConnachie, unpublished data). For the remaining 38 sites, either the weed has been washed away, or they have not been evaluated. The weevil has dispersed up to 300 km from the release sites The weed returned at 7% of the original sites within two years after control, but the weevil has been able to locate and control these infestations (McConnachie, unpublished data). This weevil has also dispersed to Zimbabwe and Mozambique unaided. Five years after the first release of S. rufinasus in South Africa, the weed no longer poses a threat to aquatic ecosystems in southern Africa. Myriophyllum aquaticum (parrot s feather) Myriophyllum aquaticum was introduced to Africa in the early 1900s (Jacot Guillarmod 1979). Unlike the other four aquatic weeds, M. aquaticum is rooted to the substrate and can grow in water up to 1.5 m deep, with emergent shoots which protrude some mm above the water surface. Outside of its native range, propagation is entirely vegetative due to the absence of male plants (Henderson 2001). The plants grow throughout the year in tropical and subtropical regions of the world. In the more temperate areas, the emergent shoots die back due to frost damage, but sprout from nodes during the spring (Cilliers 1999). Problems caused by M. aquaticum are similar to those caused by A. filiculoides, which include a reduction in stream flow, blocking of pumps and interference with recreational activities (Chickwenhere 2001). Herbicides are not translocated well down the stems of the plants and its rate of growth makes mechanical control impractical. Therefore a biological control program was initiated in South Africa in 1991, which resulted in the release of the leaf-feeding beetle, Lysathia sp. (Chrysomelidae) in late 1994 (Cilliers 1999). This insect has established at more than 25 localities throughout South Africa. The damage caused by the beetle resembles that of frost with a die-back of the emergent vegetation. However, regrowth occurs from the submerged stems. After several years of defoliation, the mat of the weed collapses and no regrowth occurs (Cilliers 1999). However, an additional agent, Listronotus marginicollis (Hustach) (Coleoptera: Curculionidae), the larvae of which bore into the stems of M. aquaticum, is being considered for release. It is hoped that this species will reduce the time taken to achieve biological control. Salvinia molesta (salvinia) Salvinia molesta is a free-floating fern that inhabits still and slow-moving freshwater bodies across the world. This fern is sterile and reproduces by vegetative growth of the rhizomes (Forno & Julien 2000). The negative impacts caused by this weed are the same as those for other floating aquatic weeds, where dense mats constrain the utilisation of impoundments and rivers for agricultural, recreational and conservation purposes (Cilliers 1991). S. molesta became a major problem on Lake Kariba as it was filling in the late 1950s. By 1962, 371

3 Proceedings of the XI International Symposium on Biological of Weeds some 1000 km 2, or 22% of the lake was covered by a thick mat of the weed (Mitchell & Rose 1979). Salvinia molesta can fairly easily be controlled with the use of herbicides. However, concerns over the use of chemicals in the aquatic environment prompted the search for a more sustainable control option. Several agents have been released against S. molesta in Africa. The weevil Cyrtobagous salviniae Calder and Sands has been the most successful and has now been released in many countries in Africa (Julien & Griffiths 1998). The adults feed on the growth tips of S. molesta, stunting its vegetative growth. The larvae feed on the buds and the roots and then burrow into the rhizome of the plant, causing the plants to rot and sink (Julien et al. 1987). Although this insect is not a particularly good disperser (Forno & Julien 2000), it has been a successful biological control agent wherever it has been introduced in the world. Salvinia molesta is under complete biological control in Africa and no longer requires any manual removal or herbicide application (Cilliers 1991). In addition, a congeneric species, Cyrtobagous singularis Hustache, was released in Botswana and Zambia in the 1970s with little success (Julien & Griffiths 1998). The moth Samea multiplicalis (Guenée) (Pyralidae) was released on Lake Kariba in 1970, but did not establish (Mitchell & Rose 1979). The grasshopper Paulinia acuminata (DeGeer) (Paulinidae) was released in Zimbabwe in 1969 and Zambia in 1970 (Julien & Griffiths 1998). This agent was damaging on S. molesta on Lake Kariba and by 1973 the area covered by the weed had been reduced to 5% and by 1980 S. molesta covered less than 1% of the dam, a situation which has been maintained since then (Mitchell & Rose 1979). Although there has been some dispute as to the efficacy of the grasshopper in controlling S. molesta, it certainly did play a role in the control of this weed on Lake Kariba, although this was most certainly in conjunction with a reduction in the nutrient status of the water body as the dam settled (Mitchell & Rose 1979). Pistia stratiotes (water lettuce) Pistia stratiotes is a rosette-type plant that floats freely on still or slow-moving water bodies. The undersides of the leaves contain spongy parenchyma and the leaves are covered with dense hairs on both surfaces. This plant reproduces through the formation of stolons and daughter plants (Forno & Julien 2000). The role of sexual reproduction is considered less important than the vegetative reproduction, although seed germination is an important factor in the dynamics of water lettuce populations (Dray & Center 1993). The negative impacts caused by this weed are the same as those for the other aquatic weed species, where dense mats limit all aspects of utilisation of rivers and dams. Once again, this weed was targeted for biological control as it was deemed the most suitable control option and the Brazilian weevil, Neohydronomus affinis (Hustache), which was previously incorrectly referred to as Neohydronomus pulchellus Hustache was introduced via Australia to a number of countries in Africa (Julien & Griffiths 1998). The larvae form extensive mines in the leaves, sometimes severing the leaf from the plant, while the adults chew characteristic round feeding holes in the leaves. Since the mid-1980s, this agent has been released in seven countries in Africa, where is has been very effective at controlling the weed. Eichhornia crassipes (water hyacinth) Eichhornia crassipes has been the most damaging weed in waterways in Africa. It has been present on this continent since it was first recorded in Egypt in the late 1800s (Gopal 1987) and South Africa since the early 1900s (Cilliers & Neser 1991). The impact of this weed on aquatic ecosystems has been staggering, and none more so than on Lake Victoria, East Africa. In the mid- 1990s, the lake was infested by up to 20,000 ha of E. crassipes, which floated around in huge mats and infested inlets and fishing beaches along the shoreline (Moorhouse & Albright 2002). It was estimated that some 80% of the Ugandan shoreline was infested with a permanent fringe of the weed extending out to around 10 m (Ogwang & Molo 1999). The proliferation of water hyacinth on Lake Victoria has been linked to an increase in the eutrophication of the water due to changes in land-use practices in its catchment (Bright 1998). The dominant socio-economic activities around Lake Victoria and its catchment, which include agriculture, hydroelectric power generation, fisheries, lake transport, recreation and water supply for both domestic and industrial use, were heavily impacted by E. crassipes. Trade in and out of the ports and in turn the productivity of the three countries surrounding the lake (Kenya, Uganda and Tanzania) were seriously affected (Hill 1999b). Since the first introductions in Zambia in 1971, five arthropod species and one pathogen species have been released against E. crassipes in Africa, with varying levels of success (Julien & Griffiths 1998). However, of these, it has been two weevil species, Neochetina eichhorniae Warner and N. bruchi Hustache, that have been credited with most of the success in the biological control of this weed (Julien et al. 1999). The first introductions of these weevils were made to Lake Victoria in By 1998, there was a significant reduction in the extent of the weed on the Ugandan shoreline and by 372

4 Technology transfer of aquatic weed biocontrol December 1999, some 75% of the mats on the Kenyan side of the lake had sunk (Anon. 2000). Between February 1998 and February 2001, the percentage coverage of E. crassipes on Lake Victoria had been reduced from about 20,000 ha to below 2000 ha (Moorhouse & Albright 2002). There has been much speculation regarding the causes for this dramatic crash on Lake Victoria and some authors (e.g. Moorhouse & Albright 2002) have cited physical factors such as the El Niño phenomenon during which the water level in the lake rose some 3 m. While there is little doubt that the increase in wind and wave action would have promoted mortality of the E. crassipes plants, it was the action of the weevils, which broke up the tight structure of the mats, that allowed the increased turbulence on the lake to further destroy the plants. The biological control of E. crassipes has also been highly successful in a number of other countries in Africa, notably Benin, where the biological control of this weed is estimated to have resulted in an increased in income of some US$ 30.5 million per year. This translates to a benefit to cost ratio of some 124:1 (De Groot et al. 2003). However, this study did not attempt to quantify the benefit to biodiversity due to reduction of the mats of water hyacinth. South Africa has had an active biological control program on E. crassipes since the mid-1980s and has released the highest number of species against this weed (Hill & Cilliers 1999). However, the results from this country have been variable and ascribed to a temperate climate, characterised by cold winters, eutrophic water bodies and interference from other control actions such as herbicide application and manual removal of the mats (Hill & Olckers 2001). This has prompted several courses of action, including surveys for additional agents which might be better cold adapted, inundative releases of agents already released at the start of spring, implementing stricter water quality guidelines, and an attempt to better integrate alternative control options with biological control (Hill & Olckers 2001). Discussion The biological control of the five aquatic weeds, with few exceptions, has been highly successful in Africa, to the point where, provided the water body is correctly managed and biological control is implemented, none of these weeds should present problems. The success of these programs on this continent can be ascribed to several factors: the reliance on fundamental research performed on the target biocontrol agents in developed countries obviating the need to screen agents for host specificity in resource-poor countries, thereby increasing the speed with which new agents can be introduced the development of simple mass-rearing techniques, ensuring the release of high numbers of healthy insects standard post-release monitoring techniques allowing comparison between different control sites in different countries the involvement of dedicated individuals who understood the potential of biological control and who ensured that the projects progressed. This is probably the single most important factor, as without these individuals stationed in research institutes throughout Africa, the concept of biological control would not have been communicated to affected riverine communities and politicians would not have been convinced as to the potential of this technique. Further to the involvement of biocontrol researchers in situ in these countries, biocontrol scientists from research institutes in the developed world, with the aid of various funding agencies were pivotal in the transfer of appropriate technology community involvement was an important factor in the success of these projects, in particular for mass rearing and agent distribution. The key to the success of any biocontrol program in Africa is that the technology that is transferred must be appropriate to the situation and that all programs must be flexible. No biological control program will succeed unless it has political support. This support has been engendered through the publicizing of success, where impacts can be observed at the landscape level and real benefits accrue to affected communities. Acknowledgements The biological control programs in Africa have been funded through numerous funding agencies, too many to mention. The senior author would like to thank Rhodes University and the Plant Protection Research Institute of the Agricultural Research Council for providing some of the funding to attend the symposium. References Anonymous (2000) Lake Victoria: Against the odds. Water Hyacinth News 1, 3 7. Ashton, P.J. (1992) Azolla infestations in South Africa: history of the introduction, scope of the problem and prospects for management. Water Quality Information Sheet. Department of Water Affairs and Forestry, Pretoria, South Africa. Bright, C. (1998) Life out of bounds. Biological invasion in a borderless world. The Worldwatch Environment Alert Series. W.W. Norton & Company, New York. Buckingham, G. (1994) Biological control of aquatic weeds. Pest management in the subtropics. Biological control a Florida perspective (eds D. Rosen, F.D. Bennett and J.L. Capinera), pp Intercept, UK. Chikwenhere, G.P. (2001) Current strategies for the management of water hyacinth on the Manyame River system in Zimbabwe. Biological and integrated control of water hyacinth, Eichhornia crassipes. Proceedings of the Second Global Working Group Meeting for the Biological and Integrated of Water Hyacinth. ACIAR Proceedings No. 102 (eds M.H. Julien, M.P. Hill, T.D. Center & Ding Jian- 373

5 Proceedings of the XI International Symposium on Biological of Weeds qing), pp Australian Centre for International Agricultural Research, Canberra. Cilliers, C.J. (1991) Biological control of water fern, Salvinia molesta (Salviniaceae), in South Africa. Agriculture, Ecosystems and Environment 37, Cilliers, C.J. (1999) Biological control of parrot s feather, Myriophyllum aquaticum (Vell) Verdc. (Haloragaceae), in South Africa. African Entomology Memoir 1, Cilliers, C.J. & Neser, S. (1991) Biological control of water hyacinth, Eichhornia crassipes (Pontederiaceae), in South Africa. Agriculture, Ecosystems and Environment 37, DeGroot, H., Ajuonu, O., Attignon, S., Djessou, R. & Neuenschwander, P. (2003) Economic impact of biological control of water hyacinth in southern Benin. Ecological Economics 45, Dray, F.A. & Center, T.D. (1993) Biocontrol of waterlettuce (Pistia stratiotes). Proceedings of the 27 th Annual Meeting, Aquatic Plant Research Programme. pp Forno, I.W. & Julien, M.H. (2000) Success in biological control of aquatic weeds by arthropods. Biological control: Measures of success (eds G. Gurr & S. Wratten), pp Kluwer Academic Publishers, The Netherlands. Gopal B. (1987) Water Hyacinth. Elsevier, New York. Gratwicke, B. & Marshall, B.E. (2001) The impact of Azolla filiculoides Lam. on animal biodiversity in streams in Zimbabwe. African Journal of Ecology 38, 1 4. Heard, T.A. & Winterton, S.L. (2000) Interactions between nutrient status and weevil herbivory in biological control of water hyacinth. Journal of Applied Ecology 37, Henderson, L. (2001) Alien weeds and invasive plants. A complete guide to declared weeds and invaders in South Africa. Handbook No. 12. Plant Protection Research Institute. Hill, M.P. (1998a) Herbivorous insect fauna associated with Azolla species in southern Africa. African Entomology 6, Hill, M.P. (1998b) Life history and laboratory host range of Stenopelmus rufinasus Gyllenhal (Coleoptera: Curculionidae), a natural enemy for Azolla filiculoides Lamarck (Azollaceae) in South Africa. Bio 43, Hill, M.P. (1999a) Biological control of red water fern, Azolla filiculoides Lamarck (Pteridophyta: Azollaceae), in South Africa. African Entomology Memoir 1, Hill, M.P. (1999b) The world s worst aquatic weed. Pesticide Outlook 10, Hill, M.P. & Cilliers, C.J. (1999) A review of the arthropod natural enemies, and factors that influence their efficacy, in the biological control of water hyacinth, Eichhornia crassipes (Mart.) Solms-Laub. (Pontederiaceae), in South Africa. African Entomology Memoir, 1, Hill, M.P. & Olckers, T. (2001) Biological control initiatives against water hyacinth in South Africa: constraining factors, success and new courses of action. Biological and integrated control of water hyacinth, Eichhornia crassipes. Proceedings of the Second Global Working Group Meeting for the Biological and Integrated of Water Hyacinth. ACIAR Proceedings No. 102 (eds M.H. Julien, M.P. Hill., T.D. Center & Ding Jianqing), pp Australian Centre for International Agricultural Research, Canberra. Holm, L.G., Plucknett, D.L., Pancho, J.V. & Herurger, J.P. (1977) The world s worst weeds, distribution and biology. The University Press of Hawaii. Jacot Guillarmod, A. (1979) Water weeds in southern Africa. Aquatic Botany 6, Julien, M.H., Bourne, A.S. & Chan, R.R. (1987) Effects of adult and larval Cyrtobagous salviniae on the floating weed, Salvinia molesta. Journal of Applied Ecology 24, Julien, M.H. & Griffiths, M.W. (1998) Biological control of weeds. A world catalogue of agents and their target weeds. Fourth Edition. CABI Publishing, UK. Julien, M.H., Griffiths, W.M. & Wright, A.D. (1999). Biological control of water hyacinth. Australian Centre for International Agricultural Research, Canberra. Lumpkin, T.A. & Plucknett, D.L. (1982) Azolla as a green manure: Use and Management in crop production. Westview Tropical Agriculture Series No. 5. Westview Press, Boulder, Colorado. Mitchell, D.S. & Rose, D.J.W. (1979) Factors affecting fluctuations in extent of Salvinia molesta on Lake Kariba. PANS 25 (2), Moorhouse, T. & Albright, T. (2002) Plotting water hyacinth s downfall. Water Hyacinth News 6, 1 3. Ogwang, J.A. & Molo, R. (1999) Impact studies on Neochetina eichhorniae and Neochetina bruchi in Lake Kyoga, Uganda. Proceedings of the 1 st IOBC Global Working Group Meeting for the Biological and Integrated of Water Hyacinth (eds M.P. Hill, M.H. Julien & T.D. Center), pp Oosthuizen, G.J. & Walters, M.M. (1961) of water fern with diesoline. Farming in South Africa 37,