Landcare Research - Manaaki Whenua

Landcare-Research -Manaaki Whenua

New Project for Vanuatu

The highly effective broom weed beetle

The highly effective broom weed beetle

Vanuatu is a beautiful, mountainous archipelago of volcanic origin in the South Pacific. Consisting of over 80 islands, and with a total land area of 12,200 km2, Vanuatu has a tropical climate moderated by southeast trade winds, and a population of just under 274,000. As a developing country Vanuatu faces its share of challenges, including top ranking in the world for natural disaster risk. Natural hazards include tropical cyclones from January to April, volcanic activity, earthquakes, tsunamis, flooding and drought. In 2015 Vanuatu suffered widespread devastation from Severe Tropical Cyclone Pam, one of the worst-ever cyclones to hit the region, and in 2017 volcanic activity required the residents of one island (Ambae) to be evacuated for several months. The economy relies on tourism and natural resources, including hardwood forests and fish, plus exporting produce, particularly kava, copra and beef.

Like most Pacific islands, Vanuatu has its share of undesirable invasive weeds, which have an impact on agricultural and horticultural activities, threaten biodiversity, and create human and animal health issues. Vanuatu also has aquatic weeds that reduce access to water bodies and increase the risk of mosquito-borne viruses, such as dengue fever, by reducing water quality and flow. Since weeds thrive on disturbance, they are often the first species to regrow following natural disasters such as cyclones. Climate change is therefore expected to further exacerbate weed issues in the region.

Most weeds are currently managed by slashing or mowing, since herbicides are expensive or not readily available. Fire is also sometimes used as a control tool, but it is not always an option, such as where grazing lands are also used for plantation crops like coconuts. Controlling weeds manually is time-consuming and the benefits are often short-lived, since in a tropical climate weeds regrow rapidly. More sustainable and cost-effective weed management options are needed.

Biocontrol is the only feasible, long-term, safe control option for most weeds once they are widespread, especially in developing countries, which have fewer resources available for pest management. Vanuatu has long embraced the use of biological control for weeds, releasing the first agent, a lacebug (Teleonemia scrupulosa), against lantana (Lantana camara) in 1935 (see table). Since then nine biological control agents have been introduced against eight weed species. “Seven of these agents have established, plus another six have arrived without assistance,” said Michael Day, of Biosecurity Queensland, who has helped to release a number of biocontrol agents in Vanuatu recently.

The most successful agent is a beetle (Calligrapha pantherina), which now provides complete control of broom weed (Sida acuta and Sida rhombifolia). Control of water hyacinth (Eichhornia crassipes) by two weevils (Neochetina bruchi and N. eichhorniae), and water lettuce (Pistia stratiotes) by a third weevil (Neohydronomus affinis), has also been fairly good in most areas. Anecdotal evidence suggests that a rust (Puccinia spegazzinii) is having an impact on mile-a-minute (Mikania micrantha), and a moth (Epiblema strenuana) has also reduced the threat posed by parthenium (Parthenium hysterophorus). However, there is considerable potential to build on these good foundations, and an ambitious new 5-year project, which began in July 2018, is aiming to do just that.

The latest project is funded by New Zealand’s Ministry of Foreign Affairs and Trade (MFAT) and is focused on key pasture weeds affecting the beef industry. Vanuatu is the largest producer of beef in the South Pacific, but a recent study found that cattle numbers and carcass weights have fallen since 2007. Pasture quality is thought to be a contributing factor, as about 90% of beef farmers report having a problem with unpalatable/poisonous weeds outcompeting more desirable pasture species. If key pasture weeds can be better managed through biocontrol, it could benefit beef farmers – and their communities – in many ways. They would have more time and money to put into other farming activities and bring more land into production. The harmful effects of unregulated herbicide use would also be reduced, especially since protective clothing and training for herbicide applicators is not often readily available.

Manaaki Whenua − Landcare Research, in collaboration with Michael Day, will work closely with Biosecurity Vanuatu on the new project. A team of five people at Biosecurity Vanuatu (Leisongi Bulesulu, Bill Garae, Jeffline Tasale, Lee Howard and Joseph Novwai) will be trained to manage the weed biocontrol programmes. Sadly, former local weed biocontrol expert Sylverio Bule passed away last year. “We are excited about working with the new team in Vanuatu and the prospect of building biocontrol expertise in the Pacific,” said project leader Lynley Hayes. “Biosecurity Vanuatu is excited to be helping cattle farmers get rid of weeds in an environmentally friendly way so they can produce quality  beef for local consumption and the export market. Ni-vans and other people love Vanuatu beef!” said Leisongi Bulesulu.

Pasture on Efate heavily infested with hibiscus burr, with some turkey berry in the background

The three worst weeds affecting pastures on the five islands most important for beef production (Efate, Epi, Malekula, Tanna and Espiritu Santo) are pico/turkey berry (Solanum torvum), hibiscus burr (Urena lobata) and wild peanut (Senna tora). No biocontrol programmes have been developed anywhere for these weeds, which can all grow to more than a metre in height. Turkey berry is a shrub that can form prickly impenetrable thickets and is readily dispersed by birds feeding on the berries. Hibiscus burr is widely grown throughout the tropics as a fibre crop, as it makes a good substitute for jute. As the name suggests, this weed has sticky burrs, which facilitate dispersal. Wild peanut dies off during the dry season but produces abundant seeds, which are readily spread by livestock. These seeds rapidly germinate once the rains return. Since all three species potentially have beneficial/medicinal uses, any potential conflicts of interest arising from biocontrol will need to be carefully evaluated.

Little is known about the natural enemies of the three novel targets, and surveys in the native range will be undertaken to determine the available options. Molecular studies are being undertaken to try to determine the origins of the Vanuatu weed populations. Since the reported native range for two of the species is vast, studies will try to pinpoint the best places to survey for potential agents. Turkey berry is reported to be native to the American tropics/subtropics, from Florida to Brazil, and hibiscus burr is reportedly native to somewhere in Asia, possibly China. Reports suggest that wild peanut is native to Belize and El Salvador, but this may be an underestimate.

An Australian project has previously looked for biocontrol agents for a very closely related, almost identical, species to wild peanut, known as sicklepod (S. obtusifolia). Sicklepod is native to South America and is problematic in Queensland. Surveys in Honduras and Mexico turned up a number of potential biocontrol agents, but the project ended before any agents could be released. Fortunately, unlike Australia, Vanuatu does not have 50 native Senna species to contend with. Some of the natural enemies of sicklepod, including four beetles and three moths, may well be sufficiently specific for Vanuatu, and appear worthy of further study as “new association agents”, in addition to any promising species found on wild peanut itself.

Other leads that will be explored for the remaining new targets include a beetle (Leptinotarsa undecimlineata), which is a close relative of a serious potato pest, the Colorado potato beetle (Leptinotarsa decimlineata), and is reported to attack turkey berry in Cuba, Honduras and Mexico. “Some no-choice testing of this beetle showed no, or at best poor, larval survival on potato, suggesting it could well be sufficiently host specific as an agent for turkey berry,” explained Quentin Paynter, who will be heavily involved in the search for new agents. Hibiscus burr is reported to be damaged by a number of pathogens in its introduced range, including one (Macrophoma urenae) that severely affected the plant when grown as a crop in Africa. The potential of these pathogens will be investigated further.

Further work is also needed to improve control of other key pasture weeds, such as nail grass / giant sensitive plant (Mimosa diplotricha), lantana and parthenium. It is uncertain whether a psyllid (Heteropsylla spinulosa) released previously against giant sensitive plant has established, and it will be reintroduced from Australia if surveys don’t find it in Vanuatu. Another agent previously released there, but not thought to have established, is a beetle (Zygogramma bicolorata) for parthenium. “The beetle was released in only small numbers, so it will be reimported from Australia and released this time in much bigger numbers,” said Michael. Lantana agents, also readily available in Australia, such as a budmite (Aceria lantanae) and/or the herring-bone fly (Ophiomyia camarae), will be introduced to Vanuatu too. So that the Biosecurity Vanuatu team can gain experience in working with a wide range of biocontrol agents, a lacebug (Carvalhotingis visenda) for the environmental weed cat’s claw creeper (Dolichandra unguis-cati) will also be introduced from Australia, and a gall mite (Colomerus spathodeae) for African tulip tree (Spathodea campanulata) from Rarotonga.

Lessons learnt from previous projects in Vanuatu and other Pacific countries are being incorporated into the new project. These include the need to build a team of experts and a wide support base for biocontrol activities, and to plan for the damage tropical cyclones can do to infrastructure and release sites. Also, there is a need to closely monitor the impact of new agents, since successful ones can work Pasture on Efate heavily infested with hibiscus burr, with some turkey berry in the background quickly in the tropics, taking years rather than decades to knock back their targets. “Although this project is initially aimed at improving the productivity of medium to large beef enterprises in Vanuatu, our hope is that it will ultimately benefit smaller enterprises, the wider pastoral sector, the environment, and other Pacific nations with similar weed
problems,” said Lynley.

We welcome any advice, information or suggestions from our international colleagues about the best places to survey for biocontrol agents for turkey berry, wild peanut and hibiscus burr, and any additional information known about their natural enemies.

Broadleaved tobacco weed
Elephantopus mollis
Tetraeuaresta obscuriventris
No, reported 1984Unknown but plant now seems rare
Broom weed
Sida spp.
Calligrapha pantherina
Yes, 2005High
Lantana camara
Teleonemia scrupulosa
Yes, 1935Slight
Uroplata girardi
Yes, 1983Slight
Calycomyza lantanae
No, reported 2012Slight
Ophiomyia lantanae
As aboveSlight
Crocidosema lantana

As above
Hypena laceratalis
As aboveSlight
Mikania micrantha
Puccinia spegazzinii
Yes, 2012Unknown but seems variable, from minor to high
Parthenium hysterophorus
Epiblema strenuana
No, reported 2014High
Water hyacinth
Eichhornia crassipes
Neochetina eichhorniae
Yes, 2004High
  Neochetina bruchi Yes, 2013 Unknown
Water lettuce
Pistia stratiotes
Neohydronomus affinis
Yes, 2006Variable

CONTACT Lynley Hayes –