Abstract
Despite being such important components of healthy functioning ecosystems, invertebrates are often overlooked in ecosystem restoration research, plants usually being the main focal point. There are many factors that can influence invertebrate communities, and mammalian disturbance is among those factors. Previous research has provided evidence that mammals can influence plant and invertebrate assemblages in a variety of direct and indirect ways. Pest-resistant fencing has been utilized as a strategy for conservation and ecosystem restoration efforts to protect areas of land on the main islands of New Zealand from introduced mammalian pests. The Orokonui Ecosanctuary (Dunedin, South Island) is a 307-hectare mainland reserve and ecosystem restoration project that utilizes pest-resistant fencing, and the vast majority of mammals have been eliminated from within the fence boundary. In the present study, the goal was to investigate how the exclusion of mammals from Orokonui using pest-resistant fencing influenced invertebrate communities by comparing invertebrate abundance and beetle diversity from sites inside the fence to sites outside the fence. How season could potentially interact with fencing was also considered. Three pairs of inside-fence and outside-fence sites for sampling invertebrates from were found, where each pair had a distinct composition of plants. Each site had their soil properties analyzed. Ground/litter-dwelling and shallow soil-dwelling invertebrates were sampled from each site, extracted using Tullgren funnels, and categorized into various taxonomic groups for abundance analyses. Beetles were further categorized down to species and recognizable taxonomic units for diversity analyses. Invertebrates deeper down in the soil had their wet weights analyzed. Soil moisture data was also collected alongside the invertebrate data and was included in the analyses as a random effect. Invertebrate and soil moisture sampling occurred once in winter and once in summer. Results showed that the abundance of several invertebrate groups differed significantly inside and outside the fence, but the differences were often not consistent between the pairs of sites. The effect of season had an interaction with the effect of the fence on the abundance of some invertebrates in some pairs. Differences in beetle diversity inside and outside the fence were also evident. No significant differences in soil invertebrate wet weights were found. Ideas for improving the present study and ideas for future research are presented. In conclusion, the exclusion of mammals using pest-resistant fencing at Orokonui does influence invertebrate communities, but the way in which the communities change can depend on the type of habitat being investigated and season.