Abstract
Invasive species are a major threat to biodiversity throughout the world. In New Zealand, invasive rats represent a significant threat to native species, especially to island seabird populations. In order to determine the ecological effects that rats have on their environment and on native species, we need information on their niche and interactions with other species. This information may then be used to inform management strategies to protect native species and control invasive rats. Furthermore, the presence and coexistence of different combinations of rodents on islands in New Zealand represents a major challenge for ecologists to explain. By investigating the mechanisms facilitating the coexistence of varying rodent combinations, we are able to improve our understanding of the interactions between rodents and how these interactions may affect other species around them.
The objective of my thesis was to identify the presence and level of interspecific competition between kiore (Rattus exulans) and ship rats (R. rattus) on Chatham Island to understand some of the mechanisms behind their coexistence and to determine the threat they pose to native species there. Specifically, I investigated the microhabitat structure and diet of kiore and ship rats in Tuku Nature Reserve on Chatham Island, New Zealand to determine whether niche partitioning may be occurring between these species. Multivariate analyses of floristic and microhabitat data using a canonical correspondence analysis and redundancy analysis, respectively, however, showed no evidence of microhabitat-niche partitioning between species. With the use of δ13C and δ15N, stable isotope analysis of kiore and ship rat tissues (liver, vibrissae and bone collagen) showed that segregation in isotopic space was occurring. I measured isotopic niche breadths using standard ellipse areas and their overlap between species for each tissue, vibrissae section and season, intraspecific isotopic variation within each species, the composition of prey sources using Bayesian mixing models, and seasonal isotopic variation between species and within species. Diet partitioning was indicated by the low isotopic niche overlap between species, by seasonal variation in the level of overlap between species, and in differences between kiore and ship rat niche breadths. However, Bayesian mixing models did not reveal major differences in diet composition between kiore and ship rats, with invertebrates, plants and the Chatham Island tāiko (Pterodroma magentae) being the most important food sources for both species. The occurrence of tāiko in the diet of both rats may either be due to direct predation or scavenging. Therefore, further studies are needed to determine whether these rats are directly impacting tāiko populations in order to prioritise eradications.
I concluded that diet partitioning was more important than microhabitat partitioning in explaining the coexistence of kiore and ship rats. However, it is possible that I was unable to detect microhabitat differences between kiore and ship rats by my methods so further studies should investigate not only sites that captured rats but those that did not to determine the microhabitat available, including sites that were not trap sites as this may have a sampling bias due to trap sites being chosen based on similar characteristics. Further mechanisms such as temporal or vertical-space partitioning may facilitate the coexistence of kiore and ship rats in Tuku Nature Reserve. The information gained here may inform pest management by determining the extent of the threat kiore and ship rats pose on the environment. Since I found both rats had broad isotopic niches, high intraspecific and seasonal variation, it is likely that these species will impact a variety of native species and can shift their diet with environmental changes, indicating that they would be able to persist in a variety of environments. Thus, pest management strategies should account for these broad and flexible niches by targeting a variety of environments for trap placements with more intensive trapping closer to vulnerable native species that these rats are likely to target as observed from my diet analysis.