Spatial complexity and microclimatic responses of epiphyte communities and their invertebrate fauna in the canopy of northern rata (Metrosideros robusta A. Cunn.: Myrtaceae) on the West Coast of the South Island, New Zealand

Rain forest canopies are renowned for their very high biodiversity and the critical role they play in key ecological processes and their influence on global climate. Despite that New Zealand supports one of the most diverse and extensive epiphyte flora of any temperate forest system, few studies have investigated epiphyte communities and their invertebrate fauna along with factors that influence their distribution and composition. This thesis represents the first comprehensive study of entire epiphyte communities and their resident invertebrate fauna in the canopy of New Zealand’s indigenous forests. The aim of this study was to determine spatial patterns of epiphyte and invertebrate species richness, abundance and community composition in relation to abiotic variables, and in particular, the responses of these communities to elevated temperature and rainfall. This study was carried out in coastal lowland podocarp-broadleaved forests at two sites on the West Coast of the South Island of New Zealand. Samples from 120 mat-forming epiphyte assemblages located on inner canopy branches of 40 northern rata (Metrosideros robusta) trees were studied to characterise the component flora and fauna. Additionally, biomass, branch and tree characteristics and community responses to treatments designed to elevate temperature and rainfall to simulate predicted climate change were measured. This investigation revealed astonishing diversity and functional complexity of epiphyte and invertebrate life in this ecosystem. The 30.6 kg (dry weight) of epiphyte material collected contained a total of 567 species, 170 epiphyte and 397 invertebrate (excluding immature specimens and mites) species, including at least 10 species new to science and many undescribed species Epiphyte communities were found to be dominated by non-vascular plants (80 % of the total species richness), particularly liverworts and invertebrate communities were dominated with respect to abundance (~ 80 % of the total individuals) by Acari, Collembola and Hymenoptera (primarily ants) and functionally by scavengers and ants. Epiphyte and invertebrate communities were highly variable with respect to spatial patterning of species richness, abundance and composition across sites, among trees within sites and among branches within trees. Overall, a highly significant proportion, > 75 %, of the variance could be attributed to differences at the branch level, but these differences could not be explained by the environmental factors measured. There were no consistent relationships between the spatial pattern of epiphytes and invertebrates, or between vascular and non-vascular plants. However, there were significant positive correlations between epiphyte biomass and invertebrate species richness (r = 0.472; p < 0.0001) and abundance (r = -0.395; p < 0.0001), as well as non-living epiphyte biomass and scavenger species richness (r = 0.4; p < 0.0001). Microclimatic measurements taken on epiphyte mats were also highly variable with respect to temperature and relative humidity at similar physical locations within the same tree as well as across trees within sites. There was also considerable variation in the intensity and frequency of climatic extremes, although potentially harmful climatic conditions were experienced by all the epiphyte mats for which weather variables were measured. Negative correlations existed between both epiphyte and invertebrate community composition and increased temperatures expressed as cumulative degree days above 5˚C. However, variability was such that there was no direct evidence that increased temperature and rainfall treatments had an effect on invertebrate species richness, abundance or diversity. Northern rata host trees harbour an astonishingly diverse and complex canopy flora and fauna that is characterised by high spatial variability. Such variability highlights that to determine species distribution and community dynamics in canopy habitats in response to disturbance caused either by climate change or invasive species the structure of entire communities at different taxonomic and spatial scales, along with their responses to microclimatic factors, need to be studied. If such complexities are not taken into account, inappropriate interpretation may result in poor decisions concerning the conservation status, vulnerability and subsequent management of such unique ecosystems.