Vegetation Recovery Following Disturbance in Alpine and Treeless Subalpine Communities in the Tasmanian Wilderness World Heritage Area

This thesis examined natural recovery on tracks in seven alpine or subalpine vegetation communities within the Tasmanian Wilderness World Heritage Area in Tasmania, Australia. In acknowledgement of the slow vegetation recovery in such communities, a chronosequence (space for time) approach was used to investigate patterns of recovery, the influence of environmental factors on such recovery and the relative resilience of the vegetation. These analyses included assessment of sites up to 23 years post trampling disturbance. The pattern of species recovery over time in the two most widespread vegetation communities, heath and sedgeland, were explored to answer the following question: What are the patterns of recovery at the species and life form level? The relative resilience (ability to recover) of these two vegetation communities 13-15 years following closure, was measured by the similarity of composition and abundance between disturbed areas and undisturbed areas. A further five vegetation communities were included in analyses to address the following questions: firstly, what is the relative resilience of these seven vegetation communities; secondly, how do environmental factors influence natural recovery; and finally, what is the relationship between (previously reported) resistance to trampling and the resilience of these communities? Further exploratory studies using paired photograph studies of tracks closed 20 years previously were used to determine the spatial patterns of recovery answering the question: what are the small-scale spatial patterns of natural recovery? Together, these data were examined to address the question of how natural recovery can be used to inform future rehabilitation or restoration programs. The present study found that there were no exclusive primary colonists in alpine or subalpine vegetation and that all colonising species occur within the adjacent undisturbed vegetation, albeit, often in lower abundance, frequency or both. Bryophytes (including mosses and liverworts) were the most frequent and abundant colonists of disturbed areas. Other strong colonists included many vegetatively reproducing graminoids such as sedges and some grasses. These species were found to largely colonise areas from lateral expansion at the edge of disturbances a phenomenon also found for other species in the spatial pattern studies. These data proved that the size, and hence edge to area ratio, can be critical in the ability of alpine and subalpine sites to recover. The high resilience of these graminoids and grasses led to greater relative resilience of communities where they are the dominant components; sedgeland and grassland. In general, this resilience also correlated well with the relatively high level of resistance in these communities which was found to be due to the dominance of flexible graminoids and grasses. However, this study also showed resistance and resilience do not always show similar responses. It was found that factors which may lead to higher resilience, such as higher available moisture, can also result in lower resistance to impacts. Environmental factors were also shown to be a major determinant of resilience of vegetation communities. This study found that sedgeland communties with mineral soils and higher compaction levels had greater persisting bare ground than those on organic soils with lower compaction. This study found that recovery in heath communities was influenced by the amount of solar radiation, with sites of lower potential, such as those on south easterly aspects, being less resilient. From the results of this study, it can be concluded that whilst recovery in alpine and subalpine areas is slow as expected, this natural recovery is highly variable between vegetation communities and environmental conditions. This differing resilience should be considered before areas are closed with the presumption they will recover or before restoration efforts are commenced. If not,l