Dr Cath Dickson (Tasmanian Land Conservancy & Affiliate with Monash University), Dr David Baker (University of Exeter), Dr Dana Bergstrom (Australian Antarctic Program), Dr Rowan Brookes (University of Melbourne), Prof Melodie McGeoch (La Trobe University ) & Dr Jennie Whinam (University of Tasmania)

As you walk across the plateau of Macquarie Island, the wind whips around you, and gaps in the clouds reveal different vistas of the striated subantarctic tundra. Together they instil a feeling of remote wilderness. The oceanic island is only 34 km long and less than 5km wide, with the Furious Fifties blowing from west to east historically maintaining the cool, wet and windy conditions.   

However, even isolated ecosystems of this World Heritage-listed island are not immune to anthropogenic pressures. Macquarie Island’s subantarctic tundra (also referred to as fellfield or feldmark) was recently described as one of 19 Australian ecological communities currently undergoing localised ecological collapse. The rapid collapse of the community is characterized by the ongoing, island-wide dieback of its dominant vascular species, the Macquarie cushion plant (Azorella macquariensis, Apiaceae). The cushion plant fulfils an important role as a foundation species on the sparse tundra-like system, where it facilitates species diversity and helps stabilise ecosystem processes (moderating fluxes in energy and nutrients). 

Dieback was first identified in the Austral summer of 2008/2009, affecting over 30% of the island-wide cushion cover. Work by a multi-agency taskforce was swift, upgrading state and federal conservation status of the Macquarie cushion plant from ‘Not Threatened’ to ‘Endangered’ and ‘Critically Endangered’ (respectively), defining special management areas, securing ex-situ collections and establishing research projects under the Federal Government’s Australian Antarctic Science Program.

Initial research hypothesised that the dieback of the cushion plant was triggered by change in climatic conditions (increased sunshine hours, wind speed and evapotranspiration). This change resulted in a reduction in plant available water during the critical summer growing season, thereby facilitating pathogenic infection of the weakened Macquarie cushions. More than ten potentially pathogenic bacterial, oomycetes, and fungal communities were identified as associated with cushion dieback. However, the cause of disease remains unknown.

Our research found that almost a decade since the first detection of disease, the island-wide dieback continues to affect nearly 30% of Macquarie cushion cover. During this time there has also been associated loss of cover, particularly in the north. There is a significant north – south gradient of cushion dieback, with dieback decreasing down the island. Advanced dieback was most prevalent in the north, and in highly exposed areas (eg Mount Tulloch) the wind had scoured diseased plants back to gravel. The most active dieback, distinguished by the yellow chlorosis of the leaves, is currently occurring in the central region of the island, while the healthiest and most extensive populations survive in the south. 

To understand the microclimate characteristics of healthy and diseased cushions, we established a network of 62 randomly stratified temperature and humidity dataloggers that were left in situ for a year. We found that diseased cushions were most prevalent at sites with less freezing events and extreme humidity. This was consistent with factors likely to promote plant pathogens (e.g. water moulds and fungal pathogens) in other taxonomic groups. Surprisingly, those variables associated with high stress (high extreme maximums, vapour pressure deficits) were unimportant in our models. This suggests that while the initial stress of extended plant water deficit likely triggered the initiation of island-wide dieback, the current conditions are now conducive to a self-sustaining pathogenic state.

Encouragingly, the most extensive and healthiest cushions occur within a natural southern refugium, where cushions are protected from disease by extremely cold (freezing) temperatures, associated with the significant north-south temperature gradient down the island. Similarly hopeful, the slow-growing cushions are also slowly recovering, as indicated by the bright spots of growth in blackened cushions, highest in the northern areas where dieback has been present for the longest period. 

Unfortunately, under predicted climate trajectories for the island, our research suggests that dieback will remain ongoing and potentially expand into those areas currently considered cold refugium. It is also clear that new areas of dieback are continuing to move through the system, affecting recovering populations and gradually eroding cushion cover over time. 

The ongoing decline of this species highlights the need for sustained recovery efforts, and to seek new solutions if we are to maintain Macquarie cushions and the associated ecosystem in their current state. As described in other recent research a clear pathway for combating ecosystem collapse requires Awareness (of the values), Anticipation (of the pressures) and (strategic) Action. Our research presented here contributes to this pathway by providing baseline information, enhancing Awareness of the value of both the species and ecological communities. We now have an increased understanding of the microclimate conditions that allows a greater Anticipation of the pressures across the island. Importantly, the new insights can be directed into Action by allowing more nuanced microrefugia to be predicted and management areas to be refined.