Using long-term population trends of an invasive herbivore to quantify the impact of management actions in the sub-Antarctic
Terauds, A and Doube, J and McKinlay, J and Springer, K, Using long-term population trends of an invasive herbivore to quantify the impact of management actions in the sub-Antarctic, Polar Biology, 37, (6) pp. 833-843. ISSN 0722-4060 (2014) [Refereed Article]
Accurate, long-term population estimates of invasive vertebrate pests are a key element of ecosystem management. Not only can they clarify the role of invasive species in changing ecosystem dynamics, they are also necessary to evaluate and assess management actions. Rabbits were first introduced to sub-Antarctic Macquarie Island in the 1870s, and since the 1960s have been targeted and influenced by a range of management programs. Here, for the first time, we model population trends of rabbits on Macquarie Island from the beginning of these management actions to the end of a recent, successful eradication attempt. We show that over a 38-year time frame, the population has undergone substantial fluctuations, peaking at over 350,000 individuals (27 indiv ha-1) in the late 1970s, before declining to less than 30,000 individuals (2-3 indiv ha-1) through the 1980s and early 1990s. From the late 1990s to 2005, the population increased relatively rapidly, this time peaking at approximately 221,000 individuals. After the commencement of eradication operations in 2010, the population dropped sharply, decreasing from 135,707 ± 25,995 to effectively zero in just over 12 months. This research contributes to our understanding of the complex population dynamics of sub-Antarctic invasive species and highlights the importance of monitoring in planning, understanding and assessing management actions. The development of models described here allowed population trends to be identified on Macquarie Island, despite 'noise' in the data from seasonality or sporadic observations. In consequence, the impacts of both long- and short-term management actions could be quantified. These techniques are applicable to other locations and species where long-term census data exist.
adaptively smoothed P-spline regression, biological invasion, census techniques, Macquarie Island, rabbits, thermal imaging