The responses of Australian butterflies to climate change

There are numerous reports of species responses to climate change over the last century and there is much interest in how species will continue to respond in the future. Much of the documentation has come from the Northern Hemisphere, where species are moving to higher latitudes and elevations, and life cycles are advancing. There has been far less research on the responses of Southern Hemisphere species to either recent or future climate change. The few studies on Australian species are biased towards vertebrates and plants with only a handful focused on terrestrial invertebrates (only one of which is a butterfly). The response of Australian species to climate change may not be consistent with those reported in the Northern Hemisphere because the environment in Australia is very different, being flat, relatively dry, with strong inland-coastal rainfall gradients and ecosystems whose dynamics are largely driven by episodic climate extremes. -- This thesis investigates the distribution and life cycle responses of Australian butterflies to recent and future climate change, using a combination of field surveys, species distribution modelling and data mining. It also investigates the accuracy of simulated species distributions that have been projected from historic data onto current climatic conditions, and whether the inclusion of host plant distributions in butterfly models improves their predictive accuracy. Investigation into range shift responses and modelling accuracy focuses on three butterfly species whereas the phenological investigation looks as a larger group of 68 species. It was found that over the last 100 years, the southern range boundaries of two of the three focal species, Elodina angulipennis and Suniana sunias, have likely remained stable, while the southern range boundary of the third species, Hypocysta euphemia, has contracted north by 2.5 degrees of latitude. The future distributions of all three species are likely to contract if the butterflies are unable to track climate change or if precipitation decreases in the future. When precipitation was projected to increase and full dispersal of the butterflies was assumed, future range expansions were indicated for two of the three species. I also found that the predictive accuracy of models projected onto another period of time is species-specific. I also found that the start of the flight season for several butterflies has changed over the last 60 years: some are emerging earlier, others later while others show no detectable trend. The phenological responses of the 68 species investigated were not associated strongly with any life history, dietary or geographic trait. To further investigate life cycle changes and monitor climate change impacts, I recommend several species as potential indicators of climate change impacts.