• Tweet this page
• Like this page
• Email this page
• Share this page

• Home»
• Publications »
• Newsletters »
• Weed Biocontrol »
• Issue 64 »
• Biocontrol for Aquatic Weeds

Biocontrol for Aquatic Weeds, a Step Closer

Leaf mining fly Hydrellia lagarosiphon. Image - University College Dublin

In Issue 56 of this newsletter we suggested that classical biocontrol had been underutilised as a method for controlling aquatic weeds in New Zealand. Currently the methods used include mechanical control, habitat manipulation, herbicides and inundative biological control using grass carp (Ctynopharyngodon idella). Mechanical methods are sometimes unavoidable (e.g. power station screen cleaners) and are expensive to implement. Herbicides are easier and cheaper to use than mechanical methods but also have limitations, particularly for submerged aquatic weeds, where there is a risk of inadequate plant exposure and uptake. There is also a risk of increasing public opposition to their use. Sterile grass carp can be an effective tool. Grass carp can be released en masse and, because they are not able to breed, can either remain in the lake until they die or be contained in cages and later removed. The downside to grass carp is that although they have plant feeding preferences, once the most palatable species are consumed they become less selective about what plant species they eat and will feed on any remaining native vegetation too. However, using grass carp to knock weeds back can at least give native plants a chance to recover once grass carp are removed or die out naturally.

Previously it was believed that biocontrol of aquatic weeds was too difficult because it was thought that most aquatic insect herbivores and pathogens have wide host ranges. However, further research has shown this not to be the case and in the past two decades there have been many examples of highly successful aquatic weed biocontrol projects overseas. After it became clear that New Zealand was missing out on a good opportunity, we organised a meeting with key stakeholders in November 2011 to explore the level of interest in developing biocontrol for aquatic weeds. Interest was high and as a result a steering group was formed (with representatives from Landcare Research, NIWA, Ministry for Primary Industries, Department of Conservation, Auckland Council, Bay of Plenty, Waikato and Greater Wellington regional councils, and Marlborough District Council), which met in June 2012 to rank potential targets (see Table) and consider how best to progress the project.

Table: Aquatic targets prioritised by key stakeholders.

The steering group acknowledged that additional funding would be required if work on aquatic weeds was to proceed, and a number of potential new funding sources were discussed. The group agreed that a detailed proposal was needed to take to potential funders. The proposal would focus only on the top three species of interest: lagarosiphon (Lagarosiphon major), hornwort (Ceratophyllum demersum) and egeria (Egeria densa). These weeds displace native vegetation, and disrupt recreational activities; storms can deposit large masses of rotting vegetation on beaches; and detached stems may block water-intakes of power stations, impeding electricity generation. All three can spread between water bodies when fragments of the plant are accidentally transferred, usually as the result of human activities such as boating and fishing. Although studies on the feasibility of biocontrol for these species had been prepared in the past, they needed to be updated in light of considerable new recent developments. Quentin Paynter has recently completed this task, and below we summarise the main findings.

Lagarosiphon is a submerged aquatic plant – native to southern Africa – that has become invasive in parts of Europe and Australia as well as in New Zealand. The plant was first recorded in New Zealand in 1950 and is now patchily distributed throughout most of the country. Lagarosiphon is dioecious, having male and female plants, but male plants are only known to occur in South Africa. The plant only reproduces asexually here and does not produce seeds. Quentin found that there are no native New Zealand plants closely related to lagarosiphon (which belongs in the Hydrocharitaceae family). “This makes it a good candidate for biocontrol as host-range testing is relatively straightforward,” said Quent. Lagarosiphon is also a problem in Irish waterways and a biocontrol project began there in 2008. A number of potential agents were identified in the plant’s native range including a leaf-mining fly (Hydrellia lagarosiphon) and a shoot-tip mining midge (cf. Polypedilum sp.). Host-range testing of these agents is well progressed and the fl y already appears to be sufficiently specific for release in New Zealand. Only a small amount of additional testing would be required to confirm this. However, more extensive testing of the midge would be required to assess its suitability for release in New Zealand.

The fly lays its eggs on vegetation growing at the water’s surface, whereas the midge has larvae that can swim, so it should be capable of attacking lagarosiphon growing in deeper water. Studies showed that the fly is capable of reducing lagarosiphon biomass by 50–70% after 134 days of exposure, and sustained herbivory over multiple generations reduced shoot biomass production by nearly 100%. The damage caused by the midge is purportedly even more harmful to the plant than that caused by the fl y. Quentin has estimated that the cost of gaining approval for the fly as a biocontrol agent for New Zealand would be in the region of $145,000–$165,000 and the cost of gaining approval for both the fl y and the midge would be ~$220,000–$260,000; pretty cheap when you compare it with the annual cost of controlling lagarosiphon in New Zealand, which was recently estimated to be nearly $1.5m.

Another Hydrellia fly looks to be a promising potential agent for egeria. Egeria (also known as Brazilian waterweed) is native to South America, but has now established in a number of countries around the world including Chile, Mexico, the United States, England, and Australia. It is most commonly encountered in the Waikato Region but occurs throughout the North Island. Egeria is at an early stage of spread in the South Island and is restricted to a small number of still, shallow and slow-moving water bodies. Surveys for potential biocontrol agents that might be suitable to release in the USA against this weed began in Argentina in 2005. The most promising species found was the Hydrellia fly, which can cause heavy defoliation. The fly has been well studied and work conducted at the USDA/ARS/South American Biological Control Laboratory indicates it has a narrow host-range, confined to the Hydrocharitaceae family. Although New Zealand has no native plants in this family, it would be prudent to test a small number of additional native New Zealand plants, but based on previous results none of these are likely to be attacked. A second Hydrellia fly has also been identified as possibly having potential against egeria but it has not been well studied yet.

Less is known about potential biocontol agents for hornwort (Ceratophyllum demersum). Hornwort (also known as coontail) has a near-global distribution and is considered native to most countries, with New Zealand being one of the few exceptions. Hornwort was thought to have been introduced as an aquarium plant from North America, and was first recorded in New Zealand in 1961. However, as part of our feasibility study, samples were collected from around New Zealand and sent to Gary Houliston who extracted DNA from the specimens and compared it with sequences from GenBank. “This indicated that hornwort in New Zealand comes from Australia and not the USA,” said Quent. NIWA considers hornwort to be New Zealand’s worst submerged weed and many other stakeholders seem to agree. Hornwort has been in the North Island for some time, and although all known infestations in the South Island have recently been successfully eradicated, this weed still presents a major threat to South Island water bodies. Given little is known about the natural enemies of hornwort, surveys would need to be undertaken, beginning in Australia and then possibly extending to other parts of the native range, depending on what is found.

The feasibility studies also address some areas of concern raised previously. Biocontrol could lead to increased physical fragmentation of aquatic weeds, which could give rise to more plants. However, in the case of lagarosiphon, this has been investigated as part of the Irish project with researchers finding that fragments damaged by the fly had significantly reduced viability. So fragmentation may not be such a big issue after all. Another potential downside to biocontrol still needs further consideration. There is potential for degraded lakes to suffer a further decline in health following the removal of weeds when they are the only remaining vegetation, and this will need to be planned for and managed through activities such as restoration of native vegetation. However, overall the prospects of achieving some really good environmental and economic outcomes relatively cheaply with aquatic weed biocontrol remain bright. Landcare Research and NIWA are joining forces to work on a detailed proposal, and it is hoped that funding can be secured soon that will allow this new collaboration, and the opportunities it provides, to be realised.

The lagarosiphon feasibility study was funded by Horizons Regional Council via an Envirolink Medium Advice Grant (1248-HZLC93), and the egeria and hornwort feasibility studies were funded by Greater Wellington Regional Council. Thanks to Jan-Robert Baars (University College Dublin) and Raymond Carruthers (USDA-ARS) for providing information about lagarosiphon and egeria respectively; Suzanne Govella (Greater Wellington), Cam Speedy (Genesis Energy), and Craig Davey (Horizons Regional Council) for providing hornwort samples for DNA work; Michelle Archer and Joe Wheeler (Mighty River Power) and Cam Speedy for information about the cost of aquatic weeds; and John Clayton (NIWA) for comments on this article.