The effect of buffel grass (cenchrus ciliaris) invasion on ant communities in central Australia

  • Sarah Bonney

    Student thesis: Other thesis - CDU


    Non-native plant invasions represent one of the most serious threats to biological diversity and ecosystem functioning worldwide. Buffel grass is an invasive perennial grass, negatively impacting ecosystems in arid Australia and internationally. Buffel grass was sown purposely to improve the grazing industry but has since spread into all mainland states of Australia. Buffel grass is known to reduce the species diversity and abundance of native plant species due to competition and development of a fire grass cycle. An increase in biomass and reduction in native plant species is believed to affect native fauna such as invertebrates, birds, reptiles and mammals although there has been limited direct evidence documented to date. In Australia, buffel grass presents a conflict between the pastoral industry and nature conservationists due to the high value
    of buffel grass and the negative impacts it has on biodiversity.

    In 2008, buffel grass was experimentally removed in small plots near Alice Springs in central Australia and native species were allowed to regenerate (B- plots). Buffel grass surrounding these plots was left unmanaged (B+ plots). In this thesis, these experimental plots were used to investigate the impact of buffel grass on ant diversity and composition. A number of questions were investigated: 1) What is the difference in the overall cover of ground vegetation between sites with and without buffel grass? 2) How do ant communities differ in sites with buffel grass to areas where buffel grass has been actively removed in experimentally manipulated sites? 3) How does the rate of seed removal vary in sites of buffel grass to areas where buffel grass has been actively removed. I hypothesize that invasion of buffel grass increases overall vegetation cover, and that this leads to a reduction in diversity of the thermophilic ant fauna, and leads to lower rates of seed removal. I also hypothesize than an increase in bare ground at sites where buffel grass was removed would be an important factor influencing ant abundance and species richness.

    Differences in ant diversity and composition were compared between B- and B+ plots. Ant abundance and species richness was also compared between two microhabitat types; bare ground and under cover, to further investigate the mechanism behind any differences between B- and B+ plots. The rate that seeds were removed by ants within B- and B+ plots was compared by placing an experimental array of seeds processing an elaiosome within plots and recording their fate.

    Ant abundance and species richness were significantly higher in B- compared with B+ plots (P = 0.007 and P = 0.001 respectively), with almost double the number of individuals and species. Species composition varied significantly between plot types with only 50% of species found in both plot types. All functional groups had a higher abundance and species richness in B- plots. This was especially pronounced for Hotclimate specialists, where total abundance was significantly higher in B- plots (P = 0.05; considering both sites together), and species richness was significantly higher at both Simpsons Gap (P = 0.009) and the Desert Park (P = 0.031). B- plots supported higher ant abundance (P = 0.064) and species richness (P = 0.005) than B+ plots on both bare ground and under vegetation cover. After 48 hrs, the mean rate of seed removal was higher in B- plots (77%) than B+ plots (63%; P = 0.028). The number of cheating incidences (consuming elaiosomes without seed dispersal) was also higher in B+ plots (41 cases compared to 29; P =0.355)

    My study has shown that the invasion of buffel grass has had a major impact on ant communities. While the increase in bare ground is the most likely mechanism, it is not a simple function of proportional increase in bare ground, but rather a broader effect of changed habitat structure. Other factors, such as food availability and interspecific competition may also be influencing ant communities. The rate of seed removal was lower at invaded sites, but how this may affect the ecosystem is not clear.
    Date of AwardJul 2011
    Original languageEnglish
    SupervisorChristine Schlesinger (Supervisor) & Alan Andersen (Supervisor)

    Cite this