Using Stable Isotope Analysis to study the diet of stoats (Mustela erminea) in the alpine zone of New Zealand

Threats posed by alien invasive species are a leading cause of global biodiversity decline. In island archipelagos such as New Zealand, introduced mammalian predators prey heavily on endemic fauna that have evolved without natural responses to this type of predation. Control of invasive species leads to positive outcomes for many threatened species but, in very large ecosystem types, conservation managers must prioritise where to place limited resources to achieve the greatest conservation gain. The New Zealand alpine zone (area above the natural timberline) makes up ~11% of New Zealand and over 30% of the conservation estate. The probability of a species being preyed upon can be seen as a combination of the numeric and functional response of a predator to their environment: the abundance of predators and the per capita consumption of the prey type in question. Stoats (Mustela erminea) are a significant predator of many alpine species, but their cryptic nature combined with the challenging working conditions of the alpine zone means no study has yet been able to apply spatial or temporal resolution to the study of stoat diet in alpine areas. As such, conservation managers currently lack information on predation feeding patterns, such as the factors that might drive greater or lesser consumption of threatened taxa by stoats. I trialled a novel technique to provide spatial and temporal replication to the study of stoat diet in alpine areas. I examined the diet of stoats (n=55) caught in the alpine zone of three national parks (NPs) of the South Island of New Zealand. In addition to examining stomach content, I conducted Stable Isotope Analysis using three stoat tissues (bone collagen, claw and liver). I examined isotopic signatures of stoat tissues against signatures of prey collected at each site (n=297 from four sites and two time periods) using Bayesian mixing model package MIXSIAR. In order to asses individual variability in diet, I also analysed the size of isotopic niches exploited by the stoats at both an individual and population level using package SIBER. Mice, ship rats and hares made up the largest portion of diet at all sites using long-term diet estimates from bone collagen. However, shorter-term estimates using claw tissue showed increased reliance on invertebrates during the summer period, supporting the hypothesis that stoats consume greater proportions of prey such as ground wētā (Hemiandrus spp) when these were seasonably more available. The proportion of birds in the diet at Fiordland and Mt Aspiring NPs was amongst the lowest recorded anywhere in New Zealand but was greater at Nelson Lakes NP, where they formed nearly a quarter of the metabolic requirements of stoats. This follows a trend seen at Nelson Lakes NP where lesser consumption of mammals has resulted in greater predation of birds, invertebrates and skinks. Ship rat abundance was strongly correlated with the proportion of mammals in the diet at all sites. Stoat abundance was positively correlated with the proportion of passerines in the diet, but stoat abundance did not explain more of the dietary variation than did site as a categorical variable. ii Stoats at Nelson Lakes NP exploited a wider range of food items than those at other sites and had a wider generalist niche. Despite this, individual stoats at this site did not have a more generalist diet and in fact were more specialised in the range of prey taken compared to other sites. This suggests individual specialisation of stoats within a generalist population niche, at least in some settings. This study is the first to assess stable isotope analysis as a tool for studying stoat diets and has produced credible results, highlighting the variability in the diets of stoats living in the alpine zone. However, uncertainty remains as to the accuracy of estimated ‘trophic enrichment factors’ utilised in models, and how these might affect results. While this study provides proof of concept that stable isotope analysis can provide useful information to inform management of invasive species in New Zealand, further research is needed to determine stoat specific enrichment factors, and to confirm the accuracy of indicative results presented here. Conservation managers should be aware of the potential for sudden and marked increases in per capita consumption of threatened alpine species by stoats. This is possibly linked to changes in abundance of mammalian prey such as ship rats. This is the first evidence to demonstrate individual specialisation within the diet of an introduced mammal at a landscape scale. This might explain accounts of episodic predation on threatened taxa and can be incorporated into management practices that aim to minimise the effects of invasive predators on species of conservation interest