Location: Office of International Research Engagement and Cooperation
2023 Annual Report
Objectives
Objective 1: Discover, identify and initiate the development of new biological control agents for invasive weeds that are native to Australia and/or Asia such as Melaleuca, Lygodium, Acacia, Hydrilla, Nymphoides, Azolla and Rotala. Establish high-priority agents by collecting, rearing, and shipping to U.S. collaborators. (non-hypothesis driven) [NP304, C2, PS 2A 2B]
Sub Objective 1.1. Discovery and identification of new biological control agents for invasive weeds and developing understanding of basic life history of natural enemies to design rearing methods and host range tests. Ship to U.S. collaborators.
Sub Objective 1.2. Development of new biological control agents for invasive weeds. Conduct host specificity tests and assess impact of potential biological control agents.
Objective 2: Develop an iterative approach in exploration for biological control agents using a decision process based on Next Generation Sequencing of weed targets in their native and introduced ranges and their natural enemies. (potentially hypothesis driven) [NP304, C2, PS 2B]
Objective 3: Selection of biological control agents compatible with chemical and mechanical control or for areas where alternative controls of weed invasions are not feasible. (non-hypothesis driven) [NP304, C2, PS 2C]
Approach
Objective 1: Develop understanding of basic life history of natural enemies to design rearing methods and host range tests. Conduct host specificity tests and assess impact of potential biological control agents. Foreign exploration will be guided by the center of origin of the weed species (classical and molecular diagnostics), habitat diversity, climate matching, specific phenology and ecology of the targets. Sites for exploration will be identified through the use of herbarium records, local knowledge, and maps of ecological areas. Exploration will be conducted in a systematic manner visiting unique ecological areas in the distribution of targets. Sites will be surveyed at different times of the year, i.e. wet and dry seasons. Agents will be sought which supplement the modes of attack of agents already established in the U.S. Initial molecular screening with mitochondrial markers will be conducted on representative herbivore species collected and the weed target species to provide an indication of genetic diversity and structuring. For selected arthropods the number of life stages, feeding behavior, developmental time of immatures, sex ratio, oviposition behavior of adults, fecundity and adult longevity will be determined in the laboratory and supplemented with field observations. Preliminary host range tests will be designed to predict the field host range of a candidate biological control agents. Studies will also be undertaken to quantify the impact of promising agents for select target species.
Objective 2: The prospective and iterative approach tests if provenance determination of the target weed can ensure there is no mismatch between areas of the native range surveyed for herbivores and the introduced ranges in terms of the target host plant. Provenance of the agent (controlling for cryptic species) is an important predictor of efficacy and host specificity of an agent.
Objective 3: Investigation of existing data on the biology/host specificity/impact of natural enemies in large data sets will lead to the identification of specific traits that influence the compatibility of agents to produce increased efficiency where alternate or existing controls for target weeds are ineffective. Large data sets on the biology, impact and host specificity of insect herbivores as well as the phenology/demography have been accumulated for weed targets over past decades by our laboratory. Contrasting this information with the weed target’s preferred habitat/s in the invasive range, current success and application of alternative forms of control (chemical/mechanical) and areas where existing released biological control agents have been ineffective, will be used to guide prioritisation of the selection of new biological control agents to be evaluated in Australia or introduced into U.S. quarantine.
Progress Report
Objective 1. New biological control agents were evaluated in Australia for invasive weeds targeted for management in the United States.
Earleaf acacia, Acacia auriculiformis (Fabaceae) is invasive in Florida, and evaluation of potential biocontrol agents continued to focus on two agents. Culturing the gall wasp, Trichilogaster sp. in a greenhouse continued to supply wasps to US quarantine as needed. Gall wasp surveys in northern Queensland were carried out in August and November 2022 and February and May 2023. Quarterly surveys will continue for the coming year and will be completed in early 2024. During 2022/23, preliminary host range testing of Macrobathra callipetala was completed to assess the likelihood of its safety as a biological control agent. No-choice host range tests were conducted and results from these tests were promising. Choice tests were completed by releasing two newly emerged adult pairs into a cage with a plant of Acacia auriculiformis and a plant of Acacia mangium. Early results indicate that adult oviposition and larval development occurred equally on both species of Acacia. In the field, eggs of the chrysomelid beetle (Calomela intemerata, currently in US quarantine) have been found inside damaged leaves tied together by M. callipetala.
Cogon grass, Imperata cylindrica, (CG) is an invasive grass in the U.S. in the Southeast but has been reported to exist as far north as West Virginia and Oregon. Research focused on expanding exploratory field surveys of CG to new regions and to new local sites in Australia. Survey sites for the project now cover a 2750 km north-south geographical range across Australia. From January to June 2023, 68 CG collections have covered over 57 sites, including 39 newly surveyed sites for the period. Damage from herbivory is most often found on the basal area of CG plants. Several show potential as biocontrol agents of CG. Over 60 species have been found associated with CG so far and have been identified using a combination of molecular and morphological identification methods. Potential agents include crown-boring moths and shoot boring and shoot galling flies. Fly infestations were high, and the damage to new shoots was prolific. Preliminary host testing against sorghum and sugar cane revealed no damage to these plant species, only CG. A stem-galling midge from Indonesia, Orseolia javanica, was considered for biological control of CG in Florida and DNA barcoding is needed to elucidate whether the Australian species is a close relative that also causes elongated stem galls. Seed-feeding wasps have emerged from field-collected inflorescences. A gall wasp larva was also collected from the basal section of a young, green shoot of CG during a field survey in New South Wales. Five weevil species have been collected from CG across sites in New South Wales, Queensland and Victoria. Several armored scale, Diaspididae sp. species have been collected in the Northern Territory and Queensland. These scales are found in the crown and stress CG, stunting its growth, discoloring foliage and occasionally causing stem death. One species was colonized in the glasshouse and were observed to damage the crown and green foliage, leaving noticeable scarring on the leaves
Old World climbing fern, Lygodium microphyllum is invasive in Florida. No field surveys were conducted during this period. Further exploration efforts are being withheld until the origins of invasive US Lygodium are determined through genomics research. A colony of the defoliating noctuidae moth, Callopistria sp. from Cape York was established in 2022 and maintained. No further host range studies have been completed given that it’s now been determined that there are likely two Southeast Asian origins of Lygodium microphyllum introductions into Florida.
Mosquito fern, Azolla pinnata, is an invasive floating fern and serious aquatic weed in Northern Florida. Field surveys for biological control agents of mosquito fern was undertaken in the Northern Territory, north Queensland, and in Northern New South Wales. Surveys in north Queensland targeted the defoliating moth, Elophila sp. while the weevil Bagous clarenciensis was targeted in the Northern Territory. Weevil colonies reared on A. pinnata were longer-lived than those on A. rubra. Weevils from both provenances when reared on A. pinnata indicated that it is the preferred developmental host despite the provenance of the field collected weevils.
Exploratory research for biocontrol agents of invasive floating heart, Nymphoides peltata, invasive across many US waterways, occurred in the Republic of Korea while Chinese collaborators processed samples for nutrient analysis. Two surveys were conducted in south Korea, targeting Bagous weevils which were discovered in high numbers at two sites near Daegu and Jeonju. Specimens were collected and brought back to the laboratory for rearing trials. Observations verified that eggs are laid on submerged fruits and larvae develop to adults within, feeding on seeds. Adults feed on flowers and leaves. Field observations indicated that adults are prevalent in the early morning and are possibly nocturnal. Biweekly surveys of the two field sites are being undertaken to monitor seasonality of these weevils. A current list of herbivores collected thus far on N. peltata in China and the Republic of Korea contains 19 insect species of which 12 require further evaluation.
Monoecious and dioecious forms of the submerged aquatic weed hydrilla, Hydrilla verticillata are invasive in the U.S. Exploration in the native range for biological control agents of hydrilla was conducted in Victoria and New South Wales. Hydrilla stocks are being maintained in the laboratory to study the biogeography of plant reproductive biotypes and the phylogenies of hydrilla and associated Hydrellia flies.
Objective 2. Progress was made to develop an iterative approach to explore for biological control agents in their native and introduced ranges using a decision process based on Next Generation Sequencing of weed targets and their natural enemies. A fast throughput system of insect specimen is being developed.
Cogon grass, Imperata cylindrica, is a relatively new project. This grass has a broad, international distribution and hence is an ideal project for testing the iterative approach of exploration in biological control given that efficiency and impact can be demonstrated from the start of a project and not retrospectively. To determine the origin of Florida introductions of cogon grass, DNA has been extracted from 288 plants, mostly from Australia, but also from Africa, China, and the USA. Extractions are underway from more samples collected by collaborators in Korea, Europe and Africa. All herbivore species collected (particularly in Australia) are being analyzed for their DNA so that once the source of the Florida infestations are determined, matched insects genotypes/biotypes from these sites where these introduced types occur can be identified and tested.
A retrospective approach was also undertaken. A population genomics approach was taken to determine whether there is a genetic basis for the presence or absence of Floracarus perrepae mite galls on Lygodium microphyllum in the field in Florida, after their release in 2006. It was concluded that there was a genetic basis to the differences in mite-fern associations. These differences have likely arisen because of multiple invasions of L. microphyllum into Florida. Unexpectedly, this analysis indicated that Australia is not the origin of the Florida invasions as was currently thought. The analysis of further international material from collections was conducted to determine the true provenance of the Florida L. microphyllum. Population genomics data was obtained for individuals from across the Asian range of L. microphyllum. These results, combined with the broad understanding of the geographic range of herbivores, provide obvious directions for future research. Several geographically separate clades were identified across the native range of L. microphyllum. The distribution of the L. microphyllum lineages broadly matches the distribution of known herbivores of L. microphyllum. Experimental research on mite associations, and the Florida Lygodium microphyllum population genomics, indicate that these insects could be highly specific to the lineages. The experiments on Floracarus perrepae from different provenances showed dramatic differences in their ability to gall Florida sourced ferns.
Objective 3. Selection of biological control agents compatible with chemical and mechanical control or for Hydrilla have been selected for study areas where alternative controls of hydrilla are not feasible. Investigation of hydrilla datasets accumulated since the mid-1980s identified gaps in the geographic surveys for biological control agents of hydrilla, Hydrilla verticillata in Australia. There is potential to explore for cold tolerant herbivores in temperate regions that could be suitable for use in the northern U.S. where hydrilla is problematic and where existing agents don’t overwinter. These surveys were initiated but follow up herbivore collection surveys this year, and in 2022, were hindered once again by severe record flooding of aquatic environments in southern temperate Australia. Genomics research determined that two genotypes Hydrellia do exist in temperate and tropical regions of Australia, one collected previously in the 1980s and released in Florida, and a new species from northern Australia. The later has been colonized for preliminary impact and host range testing. This confirms that more variation does occur in Australia in leaf-mining fly populations than previous thought.
Accomplishments
1. Biological control agent specific against the weed cogon grass, Imperata cylindrica. Biological control agents are urgently required for cogon grass, Imperata cylindrica, a federally regulated noxious invasive grass with a world-wide distribution. Cogon grass is considered one of the worst invasive species in the US, causing economic and ecological damages that impact forestry, agriculture, rangelands, and natural ecosystems. Without control this invasive grass will continue to spread requiring increased management which is problematic, and time consuming, and existing mechanical and chemical control measures are expensive. Previously, no effective biological control agents were available. Scientists from the Australian Biological Control Laboratory, an ARS-associated lab, discovered a diverse range of damaging, and possibly specific, potential biological control agents that are immediately available for shipment to US quarantine facilities. This will enable stateside ARS researchers to begin testing for efficacy and safety for release to interested parties and agencies.
