TOWARDS THE IMPROVED MANAGEMENT OF PARTHENIUM WEED: COMPLEMENTING BIOLOGICAL CONTROL WITH PLANT SUPPRESSION

Parthenium hysterophorus L., Asteraceae, commonly known as parthenium weed, is an invasive weed of global significance that has become a major weed in Australia and many other parts of the world. Parthenium weed has been reported to be a significant weed of rangelands, crops, a disrupter of biodiversity in natural ecosystems and a health hazard to people and domestic and wild animals. In Australia, parthenium weed is a Weed of National Significance (WONS) and is mainly found in Queensland where it has invaded ca. 600,000 km2 of prime pasture land reducing beef production by ca. AU$100 million annually and reducing stocking rates by up to 80 %. In Australia, integrated weed management approaches are built around a classical biological control program onto which other strategies are added. To date, 11 biological control agents have been released with some of these agents becoming widely established, having a measurable impact upon weed populations. However, parthenium weed remains a major weed of concern and improvements to the integrated weed management approach are necessary. The focus of this thesis was to investigate a new strategy for integrated weed management, then to assess the resilience of the strategy under a changing climate and finally to assess its applicability in another location. To complement the biological control approach with other management tactics, appraisal of the impact of suppressive plant species (native or introduced) was undertaken under shadehouse conditions. The shadehouse studies involved two selected suppressive plants, the native Mitchell grass (Astrebella squrossa CE Hubb.) and the introduced legume, butterfly pea (Clitoria ternatea L.) with two biological control agents, a leaf and seed feeding beetle (Zygogramma bicolorata Pallister) and a stem galling moth (Epiblema strenuana Walker). The study showed that the suppressive plants to significantly suppress weed growth (as assessed by height attainment, biomass and seed production) in the absence of the biological control agents. However, this suppressive ability could be further enhanced in the presence of one of either of the aforementioned biological control agents. In addition, x-ray analysis of the seeds produced revealed that the plant growth suppression by the biological control agents and the suppressive plants together had a significant negative effect upon seed fill, indicating that management effects beyond the present generation was feasible. The interactive effects of the biological control agents and the suppressive plants upon parthenium weed growth also resulted in a greater production of biomass (fodder) by the suppressive plants. The new combined integrated weed management approach was then tested under field conditions at two contrasting sites over a two years period in central west Queensland. In the first year, at the Injune field site, the six selected suppressive plants combined well with the biological control agents present in the field to synergistically reduce the growth of parthenium weed, by between 60 to 86 %. While in the second year the results were more variable with the suppressive plants reducing parthenium weed growth by 23 to 67 % (when the biological control agents absent) or by 47 to 91% (when the biological control agents were present). When biological control agents were present, the corresponding biomass of the suppressive plants increased in comparison to that seen in plots where biological control agents had been excluded, by between 6 to 23 %. The maximum growth advantage (in the presence of biological control agents) was achieved by buffel grass (+23 %), followed by purple pigeon grass (+19 %) and bull Mitchell grass (+17 %). Similar results were found at the Monto field site where the two selected test plants, buffel grass and butterfly pea acted with the biological control agents in a synergistic fashion to bring about the growth suppression of parthenium weed by ca. 65 %. Parthenium weed plants when grown under an elevated CO2 concentration (550 μmol mol-1), produced a greater biomass, grew taller, produced more branches and seeds per plant, and photosynthesized at a greater rate, when compared to those grown at an ambient CO2 concentration (380 μmol mol-1). The winter rust (Puccinia abrupta var partheniicola Jackson Parmelee), under the elevated CO2 concentration, induced a greater disease severity (20 % more pustules) on parthenium weed leaves than recorded on plants growing under the ambient CO2 concentration (380 μmol mol-1). As a consequence of this increased disease severity, parthenium weed plants were shorter in stature, produced a lower biomass and a reduced number of branches. This was seen under both the elevated and the ambient CO2 concentration. The stem galling moth (E. strenuana) significantly reduced the height, biomass and seed production of parthenium weed plants when grown under both the ambient and the elevated CO2 concentration. In the presence of the stem galling moth, the total seed produced was 30 to 60 % less at elevated as compared to ambient CO2 concentration. The stem galling moth had a significant negative impact upon the quality of seed with about 50 % of total seed produced at elevated CO2 concentration, not filled. When grown under the ambient CO2 concentration, the stem galling moth and buffel grass together had a strong negative effect reducing growth by as much as 65 % and seed production by 54 %. However, under an elevated CO2 concentration (550 μmol mol-1) growth was decreased by 38 % while there was no effect upon seed production. Alone the effect of the elevated CO2 concentration (550 μmol mol-1) had no effect on the growth of buffel grass. A study on the distribution of parthenium weed in Pakistan revealed that the weed has rapidly spread from the northern districts to the southern districts of the Punjab Province, in the past 10 years. The persistence of parthenium weed in these hotter and dry districts suggests that the weed can survive under more extreme conditions than previously thought. The development of a CLIMEX distribution model, that took into account this increased tolerance to drier and warmer conditions predicted that the weed could still spread into many other parts of Pakistan as well as south Asia. This increased range could include the northern parts of Pakistan, parts of southern India as well as Nepal and most parts of Sri Lanka and Bangladesh. This CLIMEX distribution model satisfied the present distribution of parthenium weed both within Pakistan and also within Australia. When an irrigation scenario was added to the CLIMEX program, more parts of the southern districts of Pakistan (Indus river basin) became suitable for parthenium weed growth, as well as its biological control agent, Z. bicolorata. This expansion of the predicted range was due to irrigation producing extra moisture into the system. In addition to the weed moving north due to temperature increase, under irrigation the weed is likely to move into south of Pakistan. The CLIMEX model indicated that there are many more areas within Australia that are suitable for parthenium weed growth, both under the present and a future climate scenarios. The predicted potential distribution of the suppressive plant, buffel grass (Cenchrus cilliaris L.) throughout the world was suggested to increase, showing considerable overlapping with parthenium weed indicating that the use of this grass as a suppressive plant in the integrated management of parthenium weed will be possible in Pakistan. The importance of the findings in this study is that parthenium weed can be more effectively managed by complementing presently existing biological control strategies with suppressive plants in Australia. This approach is likely to work in to the future under changing and variable climate, especially in locations around the globe where parthenium weed is becoming a problem. In addition, the study has shown that parthenium weed is likely to expand its geographical distribution range, under both present and future climatic conditions. In Pakistan, the biological control agent, Z. bicolorata has not reached its full potential range and is likely to undergo range expansion in the future as well. Parthenium weed is likely to become more aggressive in Australia and around the world, but the combined use of biological control agents and suppressive plants is likely to work effectively in many locations. Therefore along with the threats of future spread of the weed there are opportunities for better planning and management