More than 10 years after first proposing to target lagarosiphon (Lagarosiphon major) for biocontrol, MW-BSI has finally started preparing a release application for the Environmental Protection Authority (EPA). The candidate agent is a leaf-mining fly (Hydrellia lagarosiphon) whose larvae mine the leaves of lagarosiphon, particularly around the shoot tips, reducing photosynthetic capacity and biomass accumulation.
Lagarosiphon is a rooted, perennial, submerged aquatic macrophyte native to southern Africa that has invaded other regions of the world, most notably Ireland and New Zealand (NZ), but also several other European countries and Australia. Comonly known as oxygen weed, lagarosiphon was spread around the world as an aquarium plant, although large infestations cause local deoxygenation of water through changing water chemistry. In NZ, lagarosiphon forms dense and extensive underwater mats in lakes and slow-flowing rivers, outcompeting native aquatic plant species, disrupting water flows, increasing the risk of flooding, interfering with recreational activities such as fishing and boating, and reducing the aesthetic value of NZ’s beautiful lakes. Lagarosiphon is also a significant pest in our hydro lakes in the South Island as plants can block power generation equipment by clogging intakes.
Image: lagarosiphon invading a lake in the South Island
Lagarosiphon is dioecious (male and female flowers are on separate plants), and only female plants occur outside of the native range. Hence, reproduction is exclusively vegetative – plants spread via broken stem fragments which produce roots, giving rise to new plants and new infestations.
Currently available control methods in NZ include herbicides (mainly Diquat), mechanical and suction dredging and the application of weed matting to provide a shading effect. However, all of these have several disadvantages – they are costly, time-consuming, labour intensive, potentially have adverse environmental impacts and only provide a short-term solution. Biocontrol offers the potential for sustained long-term management of lagarosiphon in NZ with low environmental risk. The Hydrellia leaf-mining fly proposed for release is multivoltine, which means it has multiple generations per year, with many overlapping generations throughout the warmer months. Female flies lay eggs singly or in small clusters, primarily on emergent shoot tips of lagarosiphon. Newly hatched larvae initially feed on the small leaflets in the crown of shoot tips, before moving down the shoots to mine older leaves.
While biocontrol of floating aquatic macrophytes such as water lettuce (Pistia stratiotes) and salvinia (Salvinia molesta) have been highly successful in several regions of the world, classical biocontrol of submerged aquatic macrophytes, such as lagarosiphon, has only been attempted three times with variable success of the first two programmes initiated, and the third is too early to assess. Hydrilla (Hydrilla verticillata) was targeted for biocontrol in the USA where the weed is one of their most problematic and widespread weeds. Four insect biocontrol agents were released (two weevils and two leaf-mining flies in the Hydrellia genus). Lack of establishment, climate and biotype mismatches and other factors hampered successful control. Another leaf-mining fly in the Hydrellia genus was released as a biocontrol against egeria/dense water weed (Elodea densa synonym Egeria densa) in South Africa, and while there is some visible impact on the plants, the fly is heavily parasitised by local parasitoids specialising in native congeneric Hydrellia spp. Parasitism of biocontrol agents can be a major impediment to achieving population densities required to significantly reduce target weed populations. The third programme is against cabomba (Cabomba caroliniana) with a weevil (Hydrotimetes natans) released in Australia in 2023.
Image: lagarosiphon on a lakebed
When these concerns were highlighted, we embarked on a research project to fully evaluate the feasibility of using biocontrol as a management tool for lagarosiphon in NZ and its potential for success. A PhD student, Nompumelelo Baso, from Rhodes University in South Africa led several aspects of the research. We tested the Enemy Release Hypothesis (ERH) by comparing plant biomass, surface cover, and aquatic plant diversity in water bodies with lagarosiphon in the native range of South Africa, where the candidate biocontrol agent is present, and in NZ, where lagarosiphon was presumed to not have any significant natural enemies. The aim of this research was to provide insights into whether lagarosiphon is a serious weed here because of a lack of natural enemies to keep it in check, and that this can be reversed using biocontrol. We also conducted surveys throughout the distribution range of the weed in NZ to determine the herbivorous arthropod fauna associated with lagarosiphon in NZ. This aimed to determine the risk of parasitism of the leaf-mining fly, and the presence of any potential natural enemies already damaging the target weed. Additionally, Nompumelelo used MaxEnt Species Distribution models to assess current and future climatic suitability of lagarosiphon in NZ and mechanistic modelling to determine suitability of the NZ climate for establishment of the candidate biocontrol agent. Lastly, she assessed the propensity of female flies to deposit eggs on artificial substrates to test the adaptability of oviposition behaviour in systems where lagarosiphon shoot tips don’t reach the water surface.
The research showed that lagarosiphon has higher biomass and cover in NZ compared to the native range and that overall species richness and abundance of aquatic species is reduced in NZ compared to South Africa. Further, feeding damage to lagarosiphon and numbers of herbivores associated with the plant were higher in the native range compared to the invasive range. All these findings are consistent with the ERH, suggesting that biocontrol has the potential to be an effective tool to assist in the management of the weed in NZ.
The NZ surveys of lagarosiphon found a native aquatic moth (Hygraula nitens) feeding on and damaging lagarosiphon. This sparked a separate, more intensive study on the moth – more on this in a later issue - and although the damage can be impressive and will likely complement damage from the leaf-mining fly (should it be released), the moth is polyphagous, and thus augmenting its populations could risk higher rates of herbivory on its native host plants, potentially further disrupting native aquatic plant communities already threatened by invasive aquatic weeds. Further, the NZ surveys did not find any similar herbivore species closely related to the candidate agent. This, combined with a knowledge that no Hydrellia spp. are associated with submerged aquatic plants in NZ strongly suggest the risk of parasitism of the lagarosiphon leaf-mining fly to be very low. Without specialist parasitoids and predators present in NZ, the leaf-mining fly could reach high population densities capable of causing a decline in lagarosiphon populations.
Species distribution modelling indicated that more than 90% of NZ is suitable for invasion by lagarosiphon (dependent on suitable aquatic habitats), and that climate change will have limited impact on climate suitability for the weed. Interestingly, there was a lack of climatic overlap between the native and invaded ranges, providing evidence for lagarosiphon’s adaptability to a wide range of environmental conditions.
Degree-day modelling predicted that the leaf-mining fly could complete between 4.5 and 9.3 generations per year in NZ, dependant on differing climates throughout the weed’s distribution range. This suggests the fly could establish well and sustain viable populations in most parts of the country invaded by lagarosiphon. Nompumelelo’s research also demonstrated that female flies will lay eggs on artificial substrates if lagarosiphon is below the water surface. Hence, it will be possible to facilitate establishment of the fly in water bodies where plants don’t reach the water surface, however this is not the case for many water bodies invaded by lagarosiphon.
All these findings provide convincing evidence that biocontrol of lagarosiphon is worth pursuing. Host range testing of the leaf-mining fly was completed in Ireland in the early 2010s and included representatives of NZ’s most closely related native aquatic flora. We now have all the information and data required to build a strong case for release approval for this candidate agent for lagarosiphon. However, prior to submitting the application to the EPA, there is much work to be done on consultation and engagement with stakeholders and Māori, and conducting a full assessment of the economic, environmental and social risks, costs and benefits of biocontrol of lagarosiphon with the leaf-mining fly.
This project is funded by the National Biocontrol Collective and Manaaki Whenua – Bioeconomy Science Institute’s Strategic Science Investment Fund of the Ministry for Business and Innovation.
