In the mulgalands of south west Queensland state and transition models have been accepted as an improvement on the classic linear succession models but such models are yet to be thoroughly tested. Most research on vegetation dynamics in the mulgalands has focused on production systems. As a result there is little understanding of the ecology and biology of the flora.
Following the declaration of Currawinya National Park (28°50'S, 144''29'E) in 1992 grazing pressure was reduced by the removal of domestic herbivores and the implementation of feral animal control programs. This study reports on vegetation change resulting from this reduced grazing pressure and tests relevant models to explain the succession changes observed.
A vegetation survey was conducted in 1992. These data were used to classify the vegetation on Currawinya into 12 vegetation units and to identify representative areas for monitoring sites. Monitoring sites were established in the three major vegetation units; Tall Shrubland, Mulga/Eremophila gilesii and Euclaypt Woodland. The three grazing regimes investigated were on-park (where grazing pressure was reduced), off-park (where "traditional" grazing pressure was maintained), and exclosures (where all grazing pressure was removed). Vegetation parameters were measured each April and September from 1993 to 1996. The results showed that total grazing pressure had a significant impact on the vegetation. Increases in grass cover were generally correlated with reduced grazing pressure in all three vegetation types, and grass cover generally increased following good summer rain, hi contrast, grazing pressure had no significant impact on the unpalatable woody shrub population, litter cover or microbiotic crust cover. The dynamics of the annual vegetation differed in its response to grazing pressure between vegetation types.
Historically unpalatable woody shrubs (woody weeds) have increased in density in two of the vegetation units being monitored (Tall Shrubland and Mulga/Eremophila gilesii). These shrubs were removed in experimental plots using the same design as the monitoring program because the role of these shrubs is generally unknown. The removal of the shrubs had no major affect on the species composition, but the canopy cover of the herbaceous plants increased in both vegetation types. This indicates that there was competition between the shrubs and other vegetation but that grazing pressure had a greater influence on the rate of rehabilitation than the presence of woody shrubs.
The seed bank was investigated in the three vegetation units being monitored two years following destocking. There was no significant difference in the abundance of seeds between the two grazing regimes (on-park and off-park) although this could be attributed to the dry climatic conditions. The seed bank study revealed a suite of species that were not currently growing in the monitoring sites. As woody weed competition and grazing pressure were tested by exclusion in the field and were therefore not inhibiting the germination of the these species, I hypothesised that the climatic conditions were responsible.
To better understand the dynamics of these heterogeneous systems, the fine scale patterns within the monitoring sites were classified based on a sink/source relationship. An investigation showed that the sink-type patches exhibited the best water infiltration conditions but plant production and seed distribution was not significantly greater in these patch types. Therefore it seems that the fine scale patterns do not strictly function in terms of sinks and sources in this degraded landscape as reported in the literature.
Results reported in this thesis and information from the literature are combined to form functional group dynamic models that are vegetation type specific. These models are at a finer temporal and spatial scale than the broader state and transition model. The positive aspects of constructing models at this finer scale are discussed and areas that require further investigation are presented. Continued monitoring of these sites can add much to our ecological understanding of these systems which in turn will lead to better management of the systems for both production and conservation, while promoting National Parks as a viable and important landuse in the mulgalands.