Chickens or eggs?: The nature of the edaphic influence on vegetation succession in Tasmania

Boundaries between buttongrass moorland and forest in Tasmania are subject to much debate in the literature (Gilbert, 1959; Jackson, 1965; Jackson, 1968; Mount, 1979; Bowman, 1980; Bowman & Jackson, 1981; Brown & Podger, 1982; Mount, 1982; Bowman, Maclean, & Crowden, 1986a; Pemberton, 1989; Balmer, 1990; Brown, Balmer, & Podger, 2002). Conflict has arisen over the degree to which fire, vegetation, soil, climate and other environmental variables influence the stability of these boundaries. Transitions between buttongrass moorland and forest exist over a wide range of environmental conditions in Tasmania, including different climatic, topographic, edaphic and geological zones. Some claim that rainforest is the climatic climax community in high rainfall areas of Tasmania (Jackson, 1965; Jackson, 1968), yet the current overall distribution pattern is a mosaic of rainforest, mixed forest, eucalypt forest and buttongrass moorland (Kirkpatrick & Dickinson, 1984a). Jackson (1968) contended that succession in the absence of fire is occurring from moorland toward rainforest, through intermediate communities of eucalypt forest and mixed forest. Of particular interest to this debate is that 'firepromoting' buttongrass moorland is an early successional community, which often occurs directly adjacent to 'fire-sensitive' rainforest. Given the long history of fire in Tasmania (Macphail, 1980), it is generally accepted that fire is fundamental in explaining the distribution of forest and moorland (Gilbert, 1959; Jackson, 1965, 1968, 1977; Macphail & Shepherd, 1973; Bowman & Jackson, 1981; Brown & Podger, 1982; Jarman, Brown, & Kantvilas, 1984; Ellis, 1985, 1986; Bowman & Brown, 1986b; Jarman, Kantvilas, & Brown, 1988; Brown et al., 2002). There are alternative arguments. Vegetation boundaries can be related to topographically driven drainage conditions, where buttongrass moorland preferentially dominates infertile and poorly drained substrates at the expense of forest (Davis, 1940; Tarvydas, 1978; Pemberton, 1989). However, boundaries often occur in the absence of topographical constraints, drainage or geological boundaries. Debate focuses on the uncertainties of boundary persistence and stability. Many researchers have commented that in the absence of fire, boundaries between forest and moorland appear to be either stable (Mount, 1979, 1982), or succession is delayed (Gilbert, 1959; Jackson, 1968; Bowman et al., 1986a; Brown et al., 2002). The question arose - what is the cause of this apparent stability? The answer is somehow locked in a complex interaction between fire, soils and vegetation. It is difficult to separate which factor is driving the other. Thus, there is no easy answer to the 'chicken or egg' question. The possibilities are numerous. Suggestions from the literature will be presented in the following sections, however a more substantial review of the literature is presented in Appendix B. The primary aim of this project was to determine the nature of the edaphic influence on the succession process. This involved an investigation into the interaction between soil, vegetation and fire. This overall aim is expanded in section 1.4