Abstract
Invertebrates are the most globally abundant animal group and perform important ecosystem services like pollination, seed dispersal, and nutrient cycling. Despite all they do for our ecosystems, invertebrates are vastly underrepresented in the conservation science space, especially compared to more charismatic vertebrates. This is worrying as invertebrates face an ever-increasing number of pressures, including introduced predators and climate change, which threaten to decrease available habitats and increase competition and stress. The New Zealand alpine zone is expansive, covering 11% of the total land area (~29,482 km²) and hosts a diverse alpine invertebrate community, including spiders, butterflies, moths, wētā, grasshoppers, and giant carnivorous snails, many of which are undescribed or data deficient. Many native alpine invertebrates in New Zealand are large-bodied, flightless, and have few defence mechanisms, making them particularly vulnerable to introduced predators. Stoats and mice are resident-introduced mammalian predators in the alpine zone, but other predators like rats, cats, possums, and hedgehogs have also been found in alpine areas, albeit less frequently. Introduced predators have devastated invertebrates at lower elevations. They are known to consume invertebrates in alpine regions, but we do not know the extent to which they influence alpine invertebrate communities. In this thesis, I aimed to improve our understanding of large-bodied alpine invertebrates (which I define as ≥ 1cm in body size) vulnerable to introduced predators. I achieved this by first answering specific questions surrounding what time of year and environmental conditions promote large-bodied alpine invertebrate activity and which monitoring tools are the most effective in the alpine zone. Using that knowledge, I surveyed eight alpine locations with varying levels of predator control and mouse presence to try and understand how they influence the invertebrate community. Four of my sites were in Fiordland on the South Island of New Zealand, and the remaining four were on nearby offshore islands (Rakiura, Secretary Island, and Resolution Island). Lastly, I used species distribution models to investigate the spatial overlap of wētā and rodents currently and under two climate change scenarios in alpine areas of the South Island of New Zealand. I used these models to explore how climate change may influence alpine invertebrates, predators and their interaction.
To improve our ability to survey alpine invertebrates, I first used trail cameras to monitor invertebrate activity from early spring to autumn. I documented the activity of three common and easily identifiable taxonomic groups (beetles, wētā, and spiders) and how it was influenced by temperature, weather, and time of year to generate recommendations on when invertebrate surveys should take place. Night-time detections for all three invertebrate groups increased with temperature, where 95% of all observations fell between 5.4 and 13.0°C (the upper limit of night-time temperatures) and decreased with rising precipitation levels. Beetles were more active in spring and early summer, wētā were more active in early summer, and spiders were more active in mid to late summer. I then tested four surveying methods (pitfall traps, tracking tunnels, trail cameras, and visual surveys) in typical alpine habitats in Fiordland. Pitfall traps and trail cameras were effective tools for sampling the large-bodied ground-dwelling alpine invertebrate community. Conversely, visual surveys did not collect enough observations to be analysed, and tracking tunnels only reliably tracked wētā. Of the two successful methods (pitfall traps and trail cameras), pitfall traps caught slightly more individuals; they were more cost-effective and had fewer technical difficulties than trail cameras. As a result, pitfall traps were the most effective method.
Building upon the knowledge developed in my first two chapters, I sampled the invertebrate community at eight different sites across Fiordland and nearby offshore islands with varying levels of predator control and mouse presence. I wanted to see if current predator control, targeting stoats to protect forest birds, was effective for invertebrates or if additional predators, like mice, need to be controlled as well. I found that the presence of mice influenced invertebrate community composition, where wētā were less common at sites that contained mice. I also found that the largest 50% of invertebrates on the first day of sampling were larger at sites where mice were absent compared to sites where they were present. Lastly, I used species distribution models to predict the current and future potential range for mice, rats, and three species of wētā (Hemiandrus focalis, Hemiandrus maculifrons, and Deinacrida connectens) in a low and medium-high emissions scenario in alpine areas of the South Island of New Zealand. I also calculated the potential range overlap between the rodents and each of the three wētā species for each scenario. My models predicted that H. focalis and D. connectens (two alpine wētā) would lose around 30% of their potential range in the medium-high warming scenario. Conversely, the potential range for H. maculifrons (a more widespread wētā), rats, and mice is predicted to increase in both scenarios. Hemiandrus focalis and H. maculifrons are predicted to have more range overlap with rodents in the medium-high scenario, while D. connectens showed a slight decrease in range overlap. My models suggest that climate change may reduce the potential range for alpine species, allowing more widespread low-elevation species to expand into the alpine zone.
Mice are not currently part of any predator control programme in the alpine zone, nor are they included in the Predator Free 2050 programme. To better support our large-bodied alpine invertebrates, we should include mice in predator control management and invest in better tools to manage mice in the alpine landscape. We must also increase our inventory and monitoring of invertebrates so that we can understand if management interventions benefit all native biodiversity, including smaller organisms like invertebrates.