Understanding patterns of endemic dung beetle (Coleoptera: Scarabaeidae: Scarabaeinae) biodiversity in the Australian Wet Tropics rainforest: implications of climate change

Anthropogenic climate change is a major threat to global biodiversity, threatening many ecosystems and taxa with extinction. An understanding of the factors influencing species distribution and diversity is vital for assessing their vulnerability to climate change. This is particularly the case for tropical rainforest insects that constitute the bulk of all known biodiversity, provide many ecosystem services, and yet, remain poorly studied. Dung beetles (Coleoptera: Scarabaeidae: Scarabaeinae) are a keystone insect taxon of great ecological and economical importance. Their feeding and nesting behaviour contributes towards many ecological functions, such as the removal and recycling of animal waste. Thus, climate change induced alterations in dung beetle biodiversity have the potential to disrupt such ecosystem services, negatively impacting proper ecosystem functioning. This thesis investigates the drivers of, and predicts the impacts of future climate change on, endemic dung beetle biodiversity in the Australian Wet Tropics (AWT) rainforest. Specifically, the aims of the thesis were to: (1) derive accurate estimations of dung beetle "realised" distributions and species richness within the AWT, (2) identify and understand patterns and drivers of dung beetle biodiversity in the AWT bioregion and (3) along elevational gradients, and (4) predict the impacts of climate change on the endemic dung beetles of the AWT. Dung beetle distributions and patterns of biodiversity throughout the AWT were determined from standardised surveys and external databases. Targeted dung beetle surveys included dung–baited pitfall–traps across 20 sampling sites along four elevational gradients, between 2007 – 2009. The locality records from these surveys were supplemented with a large dung beetle database developed by Dr Geoff Monteith at the Queensland Museum. As patterns of biodiversity arise due to the overlap of species distributions, an accurate estimation of the distributional extent of each species was required. The potential distribution of a species generated by a species distribution model (SDM) is often larger than that of its realised distribution; this is widely acknowledged in the literature but is rarely accounted for. In Chapter 3 (Aim 1) I quantified SDM over– prediction and derived accurate representations of species "realised" distributions and overall AWT dung beetle species richness, by using expert knowledge to clip "potential" species distributions to well known biogeographic limits. Potential distributions were, on average, 10 times larger than the realised distributions for flightless species and 1.2 times larger for winged species. A realised species richness model was generated by summing individual realised SDMs, which attained a higher correlation between observed and predicted, subregional and local species richness and composition, compared to the potential species richness model. Using the realised species richness maps, I subsequently investigated patterns of dung beetle species richness and composition within well-known biogeographic subregions of the AWT (Chapter 4; Aim 2). Dung beetle subregional community structure has been influenced by historical rainforest refugia formed from climatic fluctuations of the Pleistocene. Subregions that maintained refugia during rainforest contraction events harboured distinct, species rich dung beetle assemblages, with subregionally endemic flightless species. Patterns of AWT β–diversity were due to species turnover, rather than nestedness, indicating species replacement between subregions. Patterns in AWT dung beetle species richness were mostly driven by a positive relationship with mammal species richness, indicating the importance of a functional trophic relationship. However, both dung beetle and mammal species richness were driven by similar responses to historical habitat stability (refugia) and climate variables associated with refugia, such as cooler temperatures and decreased seasonality. As rainforest refugia occurred at higher elevations I then investigated patterns of dung beetle biodiversity along elevational gradients using standardised sampling techniques (Chapter 5; Aim 3). Isolation, persistence, and speciation within upland refugia resulted in distinct, cool-adapted, upland species assemblages that are generally more species rich, more abundant, have greater biomass, and have narrower elevational ranges compared to lowland assemblages. Greater mammal species richness and biomass at higher elevations, also associated with rainforest refugia, contributed towards the persistence and increased abundance and biomass of upland dung beetle species by providing greater dung resources. On some mountaintops these resources were dominated by high elevation, subregionally endemic, small bodied, flightless species that competitively excluded other species, thus decreasing species richness and diversity at higher elevations. Upland species have narrow elevational ranges indicating adaptation to the cool upland conditions with intolerance to high temperatures making them particularly vulnerable to future climate change. To investigate the effects of climate change on the endemic dung beetles of the AWT (n = 70) I used SDMs to project their future distribution and population size at 10- year time steps up to 2085, using the future climate projections of the latest four Representative Concentration Pathways (Chapter 6; Aim 4). By 2085, 57 (81%) of the dung beetle species modelled are predicted to become threatened by losing ≥50% of their current distribution area. Additionally, 62 (88%) species are projected to lose ≥50% of their current population size, based on worst-case concentration pathway RCP8.5. Dung beetle species richness within the AWT is predicted to drastically decrease with current species hotspots losing up to 33 species by 2085 based on RCP8.5. These results are expected to be exacerbated by taking into consideration climate change induced mammal (dung beetle food resource providers) defaunation, as predicted by other studies. Decreases in dung beetle abundance and species richness are predicted to alter the many ecosystem services provided by dung beetles, thus negatively influencing the overall ecosystem health of the AWT rainforest. However, the results of this study also suggest that the negative impacts of climate change on dung beetle diversity can be reduced by increasing worldwide use of green technology and applying climate policies, as predicted by emission pathway RCP3-PD