TOP PADDOCK Newsletter 32 December 2003 Editor: Karen Richardson Department of Business Industry & Resource Development GPO Box 3000, Darwin NT 0801 Phone (08) 8999 2133 ISSN: 1320-727X Harvesting Floodplain Grass Seed with the airboat was terminated early because the water was too low and the depth variable, which Introduction made controlling the speed of the airboat difficult. There has been considerable interest in the Added to this was the danger of having to jump out of potential use of native floodplain grasses for the airboat into crocodile infested water to push the revegetation following control of Mimosa pigra. airboat through shallow areas and potential injuries For this to be practical, a ready supply of seed is from hitting submerged objects, including tree trunks. required, and methods needed to be developed for the harvest, cleaning, testing and storage seed The low pure seed yields obtained (Table1) show of the main floodplain grasses. wild harvesting of these species is not an economical proposition. A National Heritage Trust (NHT) Bushcare funded co-operative project between Agriculture, Weeds Table 1: Pure Seed yields from floodplain grasses Branch and Greening Australia staff was with a brush harvester. conducted during 2002 and 2003 to investigate commercial harvesting of floodplain grass seed. Grass Harvester Pure Seed Mode Yield (g/hour) Method: Hymenachne Tractor 1.3 A brush harvester was constructed to fit onto the Weeds Branch airboat to harvest seed of Hymenachne Airboat 0.8 Hymenachne acutigluma, Leersia hexandra and Pseudoraphis spinescens in standing water. Pseudoraphis Tractor 3.6 These grasses produce their seed mostly during the wet season when the stands are generally The Hymenachne seed yield from the airboat harvest accessible only by boat or airboat. was similar to that obtained in May 2000 with a cutter bar and catcher on the front of the airboat. Most of the “seed” caught is empty spikelets, and blows off after drying, during cleaning. The problems with the harvest are the poorly synchronised seed set of these grasses and the high arthropod pressure on the developing seeds. A Hymenachne inflorescence viewed under a microscope reveals a community of small mites, bugs, leaf hoppers, thrips and caterpillars. The quality of the seed harvested was poor for the Pseudoraphis, but good for the Hymenachne. The Brush Harvester Table 2: Germination of seed harvested. Grass Harvester Germin- Fresh Mode ation % ungerm- The construction of the brush harvester took inated % longer than anticipated as "new" ideas were needed to reduce the weight of the unit and to Hymenachne Tractor 38 16 account for the lack of control over harvest speed, Hymenachne Airboat 60 0 ie the speed of the airboat. Pseudoraphis Tractor 1 10 A frame was also constructed to allow the brush harvester to be used in shallow water by towing The low harvest rate would make the cost of seed with a small tractor or quad bike. prohibitive for end users. There is a need for these grasses to be grown as seed crops away from the As the floodplains dried out earlier than floodplains where the seed production may be anticipated in 2002 because of the low rainfall wet synchronized by controlling watering cutting and season in the Top End, the brush harvester could fertility. This situation would also remove the grasses not be tested on the floodplains until the 2003 dry from the arthropod pressure on developing season. inflorescences. The brush harvester was successfully tested at Arthur Cameron Middle Point on Marrakai Station towed by a Principal Agronomist Kubota tractor on Pseudoraphis on 20 May 2003, and on Hymenachne on 27 May 2003. To Be or Not to Be a Weed – that On 28 May, the floodplain brush harvester was successfully tested on Hymenachne when is the Question! Potential mounted on an airboat. The harvesting period Weediness of Bt Cotton in potentially disperse, and were categorised as waterway, cattle habitat, native bush and roadside. Northern Australia Three seed types; black seed, fuzzy seed and seed cotton were hand-planted within each of these Cotton in Northern Australia habitats in the 1999-2000 wet season (summer) to As southern Australia contemplates one of the simulate cottonseed dispersal. Additional sites were most severe droughts in history, the wet season in established in the initial dry season (DS, winter), plus northern Australia announces its progression with a second series of sites in the second year. Plant spectacular thunderstorms, lightning displays, and numbers and fruit production were recorded over a increasing cyclone watch warnings. Although the two-year period to quantify germination, survivorship, 2002-03 wet season was below average in a fecundity (boll production) and seedling recruitment number of regions, localised persistent lows as contributors to weediness. contributed to abundant refill of major river systems. 500 mm fell within 1 week at some Regular counts of plants at each site were used to stations in the Roper River catchment. This area calculate population growth, which was equated to is an example of the attractive land and water invasiveness, considered a major determinant of resources that has renewed interest in cotton weediness. A value greater than one indicates the production in northern Australia. However, this population is increasing and may become invasive of region also has existing stakeholders with an that habitat, where a value less than one indicates active interest in the allocation of natural the population is decreasing. resources; it is a mecca for fishermen, contains productive pastoral lands, and provides for traditional Aboriginal landholders. Environmental concerns over the allocation of land and water resources for irrigated agriculture, of which cotton is a major contender, are continually being raised by stakeholder groups. Researchers within the Australian Cotton CRC at Katherine, Kununurra and Broome are committed to addressing these concerns through evaluating viable and environmentally responsible cotton production systems. Weediness as a concern Northern Australia commercial cotton production systems would be based on genetically modified Bt cotton. This technology would be the foundation of IPM systems, which would effectively minimise pesticide use, but raised environmental concerns about the ecological effects of the transgene. One such concern was potential weediness, and whether the addition of the Bt gene would give an ecological advantage over conventional cotton in tropical areas. G.hirsutum evolved in tropical and sub-tropical areas, and isolated naturalised populations exist in northern Australia. So, the question was “Could the Bt gene enhance the ability for improved cultivars to become naturalised, or alter the ability of existing naturalised populations to persist?” The Office of the Gene Technology Regulator (OGTR) and Environment Australia (EA), requested that these weediness risks be evaluated prior to commercial release of Bt cotton in northern Australia. Answering the question A project commenced in 1999 to evaluate the potential for increased weediness of Bt cotton. This was done by comparing the growth and development of conventional and Bt cotton (containing combinations of CrylAc, Cry2Aa and Cry2Ab genes) in a range of habitats at Katherine (Kath), Kununurra (Knx) and Broome (Bme). These were habitats into which cotton could 3 Invasiveness was calculated for each of the two initial years of the project to allow populations to reach a more stable threshold in the second year after the initial establishment year, where seedling mortality was high. The main body of the project was concluded at the end of 2001 with preliminary results included in a submission by Monsanto to OGTR in mid- 2002 for commercial release of Bollgard II in northern Australia. The response from OGTR was contained in the document DIR012/2002 (see www.ogtr.gov.au/rtf/ir/dir012finalrarmp.rtf ) where specific regulation continued to be required for ° releases north of latitude 20 south until further The Kununurra Drain and Broome cattle yard information on weediness could be provided. sites were conducive to establishment of cotton volunteers, but there was no increase Final Report in invasiveness for transgenic cotton The final report for the initial two years of the compared to conventional cotton. weediness project was published in November 2002, and made publicly available for perusal by environmental, scientific and regulatory bodies (http://www.cotton.crc.org.au/Assets/PDFFiles/TB 3051.pdf or DBIRD Publications Section). The Field Code Changed main findings presented in the report for the original 13 sites sown in the first year include: Within each site, seedtype was the main Formatted: Bullets and factor influencing germination. Seedcotton Numbering had the lowest germination rate, attributed to higher moisture requirement for imbibition, and also emergence difficulties from the lint. This has important implications for invasiveness. Seedling recruitment from established volunteer plants would be from seedcotton only, so the number of plants establishing for subsequent generations would be relatively small. There were highly significant differences between sites for all the attributes measured. Poor establishment, growth and development Formatted: Bullets and occurred at all the native bush and roadside Numbering habitats. It is unlikely that cotton will be invasive of undisturbed natural habitats. Five sites had no plants remaining, and the Formatted: